This reverts commit e5e42ad2.

This patch came on top of another patch that introduced a regression.
Revert it before addressing the culprit.
Signed-off-by: NBartosz Golaszewski <bgolaszewski@baylibre.com>
Tested-by: NGuenter Roeck <linux@roeck-us.net>

This reverts commit d18fddff.

This patch came on top of another patch that introduced a regression.
Revert it before addressing the culprit.
Signed-off-by: NBartosz Golaszewski <bgolaszewski@baylibre.com>
Tested-by: NGuenter Roeck <linux@roeck-us.net>

'PTR_ERR(p) == -E*' is a stronger condition than IS_ERR(p).
Hence, IS_ERR(p) is unneeded.

The semantic patch that generates this commit is as follows:

// <smpl>
@@
expression ptr;
constant error_code;
@@
-IS_ERR(ptr) && (PTR_ERR(ptr) == - error_code)
+PTR_ERR(ptr) == - error_code
// </smpl>

Link: http://lkml.kernel.org/r/20200106045833.1725-1-masahiroy@kernel.orgSigned-off-by: NMasahiro Yamada <masahiroy@kernel.org>
Cc: Julia Lawall <julia.lawall@lip6.fr>
Acked-by: Stephen Boyd <sboyd@kernel.org> [drivers/clk/clk.c]
Acked-by: Bartosz Golaszewski <bgolaszewski@baylibre.com> [GPIO]
Acked-by: Wolfram Sang <wsa@the-dreams.de> [drivers/i2c]
Acked-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> [acpi/scan.c]
Acked-by: NRob Herring <robh@kernel.org>
Cc: Eric Biggers <ebiggers@kernel.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

If a driver consuming the GPIO chip is being probed at the same time as
the GPIO driver is registering the chip, it is possible for the
consuming driver to see the ->descs array in an uninitialised state.
For example, the gpio-keys-polled driver can fail like this:

    kernel: gpiod_request: invalid GPIO (no device)
    kernel: gpio-keys-polled PRP0001:07: failed to get gpio: -22
    kernel: gpio-keys-polled: probe of PRP0001:07 failed with error -22

This patch makes gpiochip_add() hold the lock protecting gpio_devices
until it has finished setting desc->gdev on the newly inserted list
entry.
Signed-off-by: NDan Callaghan <dan.callaghan@opengear.com>
Link: https://lore.kernel.org/r/20200121001216.15964-1-dan.callaghan@opengear.comSigned-off-by: NLinus Walleij <linus.walleij@linaro.org>

gpiochip_set_chained_irqchip() would assign a chained handler
to a GPIO chip. We now populate struct gpio_irq_chip for all
chained GPIO irqchips so drop this function.

Cc: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: NLinus Walleij <linus.walleij@linaro.org>
Link: https://lore.kernel.org/r/20200113220800.77817-1-linus.walleij@linaro.orgSigned-off-by: NLinus Walleij <linus.walleij@linaro.org>

In the current codes, the following 3 lines would be output to
the console for each irq line.
  gpio gpiochip0: (gpio_thunderx): allocate IRQ 10, hwirq 0
  gpio gpiochip0: (gpio_thunderx): found parent hwirq 245784
  gpio gpiochip0: (gpio_thunderx): alloc_irqs_parent for 10 parent hwirq 245784

In general, there are about tens of irq lines for each gpio chip,
and then it would emit so many insignificant log in the boot process.
These infos are more suitable for the dbg purpose. So change these
to the dbg level. With this change, about 200 lines are suppressed
on my Marvell cn96xx board.
Signed-off-by: NKevin Hao <haokexin@gmail.com>
Link: https://lore.kernel.org/r/20200120095625.25164-1-haokexin@gmail.comSigned-off-by: NLinus Walleij <linus.walleij@linaro.org>

gpio_set_config() simply call gpio_do_set_config(),
so remove the duplicated function.
Signed-off-by: NYueHaibing <yuehaibing@huawei.com>
Link: https://lore.kernel.org/r/20200116142927.58908-1-yuehaibing@huawei.comSigned-off-by: NLinus Walleij <linus.walleij@linaro.org>

drivers/gpio/gpiolib.c: In function gpio_set_config:
drivers/gpio/gpiolib.c:3053:16: warning:
 variable config set but not used [-Wunused-but-set-variable]

commit d90f3685 ("gpiolib: have a single place
of calling set_config()") left behind this unused variable.
Reported-by: NHulk Robot <hulkci@huawei.com>
Signed-off-by: NYueHaibing <yuehaibing@huawei.com>
Link: https://lore.kernel.org/r/20200108121117.45060-1-yuehaibing@huawei.comReviewed-by: NBartosz Golaszewski <bgolaszewski@baylibre.com>
Signed-off-by: NLinus Walleij <linus.walleij@linaro.org>

I see the following lockdep splat in the qcom pinctrl driver when
attempting to suspend the device.

 ============================================
 WARNING: possible recursive locking detected
 5.4.2 #2 Tainted: G S
 --------------------------------------------
 cat/6536 is trying to acquire lock:
 ffffff814787ccc0 (&irq_desc_lock_class){-.-.}, at: __irq_get_desc_lock+0x64/0x94

 but task is already holding lock:
 ffffff81436740c0 (&irq_desc_lock_class){-.-.}, at: __irq_get_desc_lock+0x64/0x94

 other info that might help us debug this:
  Possible unsafe locking scenario:

        CPU0
        ----
   lock(&irq_desc_lock_class);
   lock(&irq_desc_lock_class);

  *** DEADLOCK ***

  May be due to missing lock nesting notation

 7 locks held by cat/6536:
  #0: ffffff8140e0c420 (sb_writers#7){.+.+}, at: vfs_write+0xc8/0x19c
  #1: ffffff8121eec480 (&of->mutex){+.+.}, at: kernfs_fop_write+0x128/0x1f4
  #2: ffffff8147cad668 (kn->count#263){.+.+}, at: kernfs_fop_write+0x130/0x1f4
  #3: ffffffd011446000 (system_transition_mutex){+.+.}, at: pm_suspend+0x108/0x354
  #4: ffffff814302b970 (&dev->mutex){....}, at: __device_suspend+0x16c/0x420
  #5: ffffff81436740c0 (&irq_desc_lock_class){-.-.}, at: __irq_get_desc_lock+0x64/0x94
  #6: ffffff81479b8c10 (&pctrl->lock){....}, at: msm_gpio_irq_set_wake+0x48/0x7c

 stack backtrace:
 CPU: 4 PID: 6536 Comm: cat Tainted: G S                5.4.2 #2
 Call trace:
  dump_backtrace+0x0/0x174
  show_stack+0x20/0x2c
  dump_stack+0xdc/0x144
  __lock_acquire+0x52c/0x2268
  lock_acquire+0x1dc/0x220
  _raw_spin_lock_irqsave+0x64/0x80
  __irq_get_desc_lock+0x64/0x94
  irq_set_irq_wake+0x40/0x144
  msm_gpio_irq_set_wake+0x5c/0x7c
  set_irq_wake_real+0x40/0x5c
  irq_set_irq_wake+0x70/0x144
  cros_ec_rtc_suspend+0x38/0x4c
  platform_pm_suspend+0x34/0x60
  dpm_run_callback+0x64/0xcc
  __device_suspend+0x314/0x420
  dpm_suspend+0xf8/0x298
  dpm_suspend_start+0x84/0xb4
  suspend_devices_and_enter+0xbc/0x628
  pm_suspend+0x214/0x354
  state_store+0xb0/0x108
  kobj_attr_store+0x14/0x24
  sysfs_kf_write+0x4c/0x64
  kernfs_fop_write+0x158/0x1f4
  __vfs_write+0x54/0x18c
  vfs_write+0xdc/0x19c
  ksys_write+0x7c/0xe4
  __arm64_sys_write+0x20/0x2c
  el0_svc_common+0xa8/0x160
  el0_svc_compat_handler+0x2c/0x38
  el0_svc_compat+0x8/0x10

This is because the msm_gpio_irq_set_wake() function calls
irq_set_irq_wake() as a backup in case the irq comes in during the path
to idle. Given that we're calling irqchip functions from within an
irqchip we need to set the lockdep class to be different for this child
controller vs. the default one that the parent irqchip gets.

This used to be done before this driver was converted to hierarchical
irq domains in commit e35a6ae0 ("pinctrl/msm: Setup GPIO chip in
hierarchy") via the gpiochip_irq_map() function. With hierarchical irq
domains this function has been replaced by
gpiochip_hierarchy_irq_domain_alloc(). Therefore, set the lockdep class
like was done previously in the irq domain path so we can avoid this
lockdep warning.

Fixes: fdd61a01 ("gpio: Add support for hierarchical IRQ domains")
Cc: Thierry Reding <treding@nvidia.com>
Cc: Brian Masney <masneyb@onstation.org>
Cc: Lina Iyer <ilina@codeaurora.org>
Cc: Marc Zyngier <maz@kernel.org>
Cc: Maulik Shah <mkshah@codeaurora.org>
Signed-off-by: NStephen Boyd <swboyd@chromium.org>
Link: https://lore.kernel.org/r/20200114231103.85641-1-swboyd@chromium.orgSigned-off-by: NLinus Walleij <linus.walleij@linaro.org>

If the gpio's parent irqdomain is a msi irqdomain, we should ignore
the EEXIST error returned by the msi irqdomain because all the msi
interrupts have already been allocated.
Signed-off-by: NKevin Hao <haokexin@gmail.com>
Link: https://lore.kernel.org/r/20200114082821.14015-4-haokexin@gmail.comSigned-off-by: NLinus Walleij <linus.walleij@linaro.org>

Some gpio's parent irqdomain may not use the struct irq_fwspec as
argument, such as msi irqdomain. So rename the callback
populate_parent_fwspec() to populate_parent_alloc_arg() and make it
allocate and populate the specific struct which is needed by the
parent irqdomain.
Signed-off-by: NKevin Hao <haokexin@gmail.com>
Link: https://lore.kernel.org/r/20200114082821.14015-3-haokexin@gmail.comSigned-off-by: NLinus Walleij <linus.walleij@linaro.org>

Unduplicate the offset check by simply calling gpiochip_get_desc() and
checking its return value.
Signed-off-by: NBartosz Golaszewski <bgolaszewski@baylibre.com>
Reviewed-by: NAndy Shevchenko <andy.shevchenko@gmail.com>
Reviewed-by: NLinus Walleij <linus.walleij@linaro.org>

Unduplicate the ngpio check by simply calling gpiochip_get_desc() and
checking its return value.
Signed-off-by: NBartosz Golaszewski <bgolaszewski@baylibre.com>
Reviewed-by: NAndy Shevchenko <andy.shevchenko@gmail.com>
Reviewed-by: NLinus Walleij <linus.walleij@linaro.org>

Unduplicate the ngpio check by simply calling gpiochip_get_desc() and
checking its return value.
Signed-off-by: NBartosz Golaszewski <bgolaszewski@baylibre.com>
Reviewed-by: NAndy Shevchenko <andy.shevchenko@gmail.com>
Reviewed-by: NLinus Walleij <linus.walleij@linaro.org>

gpiochip_get_desc() takes a u16 hwnum, but it turns out most users don't
respect that and usually pass an unsigned int. Since implicit casting to
a smaller type is dangerous - let's change the type of hwnum to unsigned
int in gpiochip_get_desc() and in gpiochip_request_own_desc() where the
size of hwnum is not respected either and who's a user of the former.

This is safe as we then check the hwnum against the number of lines
before proceeding in gpiochip_get_desc().
Signed-off-by: NBartosz Golaszewski <bgolaszewski@baylibre.com>
Reviewed-by: NAndy Shevchenko <andy.shevchenko@gmail.com>
Reviewed-by: NLinus Walleij <linus.walleij@linaro.org>

Instead of calling the gpiochip's set_config() callback directly and
checking its existence every time - just add a new routine that performs
this check internally. Call it in gpio_set_config() and
gpiod_set_transitory(). Also call it in gpiod_set_debounce() and drop
the check for chip->set() as it's irrelevant to this config option.
Signed-off-by: NBartosz Golaszewski <bgolaszewski@baylibre.com>
Reviewed-by: NAndy Shevchenko <andy.shevchenko@gmail.com>
Reviewed-by: NLinus Walleij <linus.walleij@linaro.org>

Checkpatch complains about using 'unsigned' instead of 'unsigned int'.
Signed-off-by: NBartosz Golaszewski <bgolaszewski@baylibre.com>
Reviewed-by: NAndy Shevchenko <andy.shevchenko@gmail.com>
Reviewed-by: NLinus Walleij <linus.walleij@linaro.org>

There's a confusing indention in gpiochip_add_data_with_key(), which
could be misinterpreted on a quick walkthrough. Fixing this in order
to improve code readability a bit.
Signed-off-by: NEnrico Weigelt, metux IT consult <info@metux.net>
Signed-off-by: NBartosz Golaszewski <bgolaszewski@baylibre.com>

Add possibility to toggle active-low flag of a gpio descriptor. This is
useful for compatibility code, where defaults are inverted vs DT gpio
flags or the active-low flag is taken from elsewhere.
Acked-by: NLinus Walleij <linus.walleij@linaro.org>
Signed-off-by: NMichał Mirosław <mirq-linux@rere.qmqm.pl>
Link: https://lore.kernel.org/r/7ce0338e01ad17fa5a227176813941b41a7c35c1.1576031637.git.mirq-linux@rere.qmqm.plSigned-off-by: NUlf Hansson <ulf.hansson@linaro.org>

We have for some time the assign_bit() API to replace open coded

	if (foo)
		set_bit(n, bar);
	else
		clear_bit(n, bar);

Use this API in GPIO library code.

No functional change intended.
Signed-off-by: NAndy Shevchenko <andriy.shevchenko@linux.intel.com>
Link: https://lore.kernel.org/r/20191204194229.64251-2-andriy.shevchenko@linux.intel.comSigned-off-by: NLinus Walleij <linus.walleij@linaro.org>

The string literal "gpiochip" is used in several places.
Add a definition for it, and use it everywhere, to make sure everything
stays in sync.
Signed-off-by: NGeert Uytterhoeven <geert+renesas@glider.be>
Link: https://lore.kernel.org/r/20191127084253.16356-2-geert+renesas@glider.beReviewed-by: NUlrich Hecht <uli+renesas@fpond.eu>
Reviewed-by: NEugeniu Rosca <erosca@de.adit-jv.com>
Signed-off-by: NLinus Walleij <linus.walleij@linaro.org>

gpiolib has a corner case with open drain outputs that are emulated.
When such outputs are outputting a logic 1, emulation will set the
hardware to input mode, which will cause gpiod_get_direction() to
report that it is in input mode. This is different from the behaviour
with a true open-drain output.

Unify the semantics here.

Cc: <stable@vger.kernel.org>
Suggested-by: NLinus Walleij <linus.walleij@linaro.org>
Signed-off-by: NRussell King <rmk+kernel@armlinux.org.uk>
Signed-off-by: NBartosz Golaszewski <bgolaszewski@baylibre.com>

When a GPIO offset in a lookup table is out-of-range, the printed error
message (1) does not include the actual out-of-range value, and (2)
contains an off-by-one error in the upper bound.

Avoid user confusion by also printing the actual GPIO offset, and
correcting the upper bound of the range.
While at it, use "%u" for unsigned int.

Sample impact:

    -requested GPIO 0 is out of range [0..32] for chip e6052000.gpio
    +requested GPIO 0 (45) is out of range [0..31] for chip e6052000.gpio

Fixes: 2a3cf6a3 ("gpiolib: return -ENOENT if no GPIO mapping exists")
Signed-off-by: NGeert Uytterhoeven <geert+renesas@glider.be>
Link: https://lore.kernel.org/r/20191127095919.4214-1-geert+renesas@glider.beSigned-off-by: NLinus Walleij <linus.walleij@linaro.org>

Signed-off-by: NGeert Uytterhoeven <geert+renesas@glider.be>
Link: https://lore.kernel.org/r/20191120143745.1254-1-geert+renesas@glider.beSigned-off-by: NLinus Walleij <linus.walleij@linaro.org>

There should be spaces between logical operators and their operands.
Signed-off-by: NBartosz Golaszewski <bgolaszewski@baylibre.com>
Signed-off-by: NLinus Walleij <linus.walleij@linaro.org>

When IRQ chip is being added by GPIO library, the ACPI based platform expects
GPIO <-> pin mapping ranges to be initialized in order to correctly initialize
ACPI event mechanism on affected platforms. Unfortunately this step is missed.

Introduce ->add_pin_ranges() callback to fill the above mentioned gap.
Signed-off-by: NAndy Shevchenko <andriy.shevchenko@linux.intel.com>
Reviewed-by: NLinus Walleij <linus.walleij@linaro.org>
Reviewed-by: NMika Westerberg <mika.westerberg@linux.intel.com>
Reviewed-by: NHans de Goede <hdegoede@redhat.com>
Tested-by: NHans de Goede <hdegoede@redhat.com>

Add the GPIOHANDLE_SET_CONFIG_IOCTL to the gpio chardev.
The ioctl allows some of the configuration of a requested handle to be
changed without having to release the line.
The primary use case is the changing of direction for bi-directional
lines.

Based on initial work by Bartosz Golaszewski <bgolaszewski@baylibre.com>
Signed-off-by: NKent Gibson <warthog618@gmail.com>
Signed-off-by: NBartosz Golaszewski <bgolaszewski@baylibre.com>

Move validation of line handle flags into helper function.
This reduces the size and complexity of linehandle_create and allows the
validation to be reused elsewhere.
Signed-off-by: NKent Gibson <warthog618@gmail.com>
Signed-off-by: NBartosz Golaszewski <bgolaszewski@baylibre.com>

Allow pull up/down bias to be set on output lines.
Use case is for open source or open drain applications where
internal pull up/down may conflict with external biasing.
Signed-off-by: NKent Gibson <warthog618@gmail.com>
Signed-off-by: NBartosz Golaszewski <bgolaszewski@baylibre.com>

Allow pull up/down bias to be disabled, allowing the line to float
or to be biased only by external circuitry.
Use case is for where the bias has been applied previously, either
by default or by the user, but that setting may conflict with the
current use of the line.
Signed-off-by: NKent Gibson <warthog618@gmail.com>
Signed-off-by: NBartosz Golaszewski <bgolaszewski@baylibre.com>

Add support for pull up/down to lineevent_create.
Use cases include receiving asynchronous presses from a
push button without an external pull up/down.
Signed-off-by: NKent Gibson <warthog618@gmail.com>
Signed-off-by: NBartosz Golaszewski <bgolaszewski@baylibre.com>

Add pull-up/pull-down flags to the gpio line get and
set ioctl() calls.  Use cases include a push button
that does not have an external resistor.

Addition use cases described by Limor Fried (ladyada) of
Adafruit in this PR for Adafruit_Blinka Python lib:
https://github.com/adafruit/Adafruit_Blinka/pull/59Signed-off-by: NDrew Fustini <drew@pdp7.com>
[Kent: added BIAS to GPIO flag names and restrict application to input
lines]
Signed-off-by: NKent Gibson <warthog618@gmail.com>
Signed-off-by: NBartosz Golaszewski <bgolaszewski@baylibre.com>

This reverts commit 8c550e94.

This was prematurely applied and we need to back it out to merge
a better version of the development track for this feature.
Signed-off-by: NLinus Walleij <linus.walleij@linaro.org>

of_gpiochip_add(), when fails, calls gpiochip_remove_pin_ranges().

ADD:
  gpiochip_add_data_with_key() ->
    of_gpiochip_add() -> (ERROR path)
      gpiochip_remove_pin_ranges()

At the same time of_gpiochip_remove() calls exactly the above mentioned
function unconditionally and so does gpiochip_remove().

REMOVE:
  gpiochip_remove() ->
    gpiochip_remove_pin_ranges()
    of_gpiochip_remove() ->
      gpiochip_remove_pin_ranges()

Since gpiochip_remove() calls gpiochip_remove_pin_ranges() unconditionally,
we have duplicate call to the same function when it's not necessary.

Move gpiochip_remove_pin_ranges() from of_gpiochip_add() to gpiochip_add()
to avoid duplicate calls and be consistent with the explicit call in
gpiochip_remove().

Fixes: e93fa3f2 ("gpiolib: remove duplicate pin range code")
Depends-on: f7299d44 ("gpio: of: Fix of_gpiochip_add() error path")
Cc: Geert Uytterhoeven <geert+renesas@glider.be>
Signed-off-by: NAndy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: NLinus Walleij <linus.walleij@linaro.org>

The GPIO irqchip needs to initialize the valid mask
before initializing the IRQ hardware, because sometimes
the latter require the former to be executed first.

Cc: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Cc: Mika Westerberg <mika.westerberg@linux.intel.com>
Reported-by: NHans de Goede <hdegoede@redhat.com>
Link: https://lore.kernel.org/r/20191030122914.967-1-linus.walleij@linaro.orgAcked-by: NHans de Goede <hdegoede@redhat.com>
Reviewed-by: NMika Westerberg <mika.westerberg@linux.intel.com>
Reviewed-by: NAndy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: NLinus Walleij <linus.walleij@linaro.org>

After changing the drivers to use GPIO core to add an IRQ chip
it appears that some of them requires a hardware initialization
before adding the IRQ chip.

Add an optional callback ->init_hw() to allow that drivers
to initialize hardware if needed.

This change is a part of the fix NULL pointer dereference
brought to the several drivers recently.

Cc: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: NAndy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: NLinus Walleij <linus.walleij@linaro.org>

Add pull-up/pull-down flags to the gpio line get and
set ioctl() calls.  Use cases include a push button
that does not have an external resistor.

Addition use cases described by Limor Fried (ladyada) of
Adafruit in this PR for Adafruit_Blinka Python lib:
https://github.com/adafruit/Adafruit_Blinka/pull/59Signed-off-by: NDrew Fustini <drew@pdp7.com>
Link: https://lore.kernel.org/r/20190921102522.8970-1-drew@pdp7.comSigned-off-by: NLinus Walleij <linus.walleij@linaro.org>

This introduces fwnode_gpiod_get_index() that iterates through common gpio
suffixes when trying to locate a GPIO within a given firmware node.

We also switch devm_fwnode_gpiod_get_index() to call
fwnode_gpiod_get_index() instead of iterating through GPIO suffixes on
its own.
Reviewed-by: NAndy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: NDmitry Torokhov <dmitry.torokhov@gmail.com>
Link: https://lore.kernel.org/r/20190913032240.50333-3-dmitry.torokhov@gmail.comReviewed-by: NMika Westerberg <mika.westerberg@linux.intel.com>
Signed-off-by: NLinus Walleij <linus.walleij@linaro.org>

Move all the flags sanitization before any memory allocation in
lineevent_create() in order to remove a couple unneeded gotos.
Signed-off-by: NBartosz Golaszewski <bgolaszewski@baylibre.com>

When emulating open-drain/open-source by not actively driving the output
lines - we're simply changing their mode to input. This is wrong as it
will then make it impossible to change the value of such line - it's now
considered to actually be in input mode. If we want to still use the
direction_input() callback for simplicity then we need to set FLAG_IS_OUT
manually in gpiod_direction_output() and not clear it in
gpio_set_open_drain_value_commit() and
gpio_set_open_source_value_commit().

Fixes: c663e5f5 ("gpio: support native single-ended hardware drivers")
Cc: stable@vger.kernel.org
Reported-by: NKent Gibson <warthog618@gmail.com>
Signed-off-by: NBartosz Golaszewski <bgolaszewski@baylibre.com>