Commit 8eb1f71e ("gpiolib: consolidate GPIO lookups") refactors
fwnode_get_named_gpiod() and gpiod_get_index() into a unified
gpiod_find_and_request() helper.

The old functions both initialized their local lookupflags variable to
GPIO_LOOKUP_FLAGS_DEFAULT, but the new code leaves it uninitialized.

This is a problem for at least ACPI platforms, where acpi_find_gpio()
only does a bunch of *lookupflags |= GPIO_* statements and thus relies
on the variable being initialized.

The variable not being initialized leads to:

1. Potentially the wrong flags getting used
2. The check for conflicting lookup flags in gpiod_configure_flags():
   "multiple pull-up, pull-down or pull-disable enabled, invalid config"
   sometimes triggering, making the GPIO unavailable

Restore the initialization of lookupflags to GPIO_LOOKUP_FLAGS_DEFAULT
to fix this.

Fixes: 8eb1f71e ("gpiolib: consolidate GPIO lookups")
Signed-off-by: NHans de Goede <hdegoede@redhat.com>
Signed-off-by: NBartosz Golaszewski <bartosz.golaszewski@linaro.org>

While any of the GPIO cdev syscalls is in progress, the kernel can call
gpiochip_remove() (for instance, when a USB GPIO expander is disconnected)
which will set gdev->chip to NULL after which any subsequent access will
cause a crash.

To avoid that: use an RW-semaphore in which the syscalls take it for
reading (so that we don't needlessly prohibit the user-space from calling
syscalls simultaneously) while gpiochip_remove() takes it for writing so
that it can only happen once all syscalls return.

Fixes: d7c51b47 ("gpio: userspace ABI for reading/writing GPIO lines")
Fixes: 3c0d9c63 ("gpiolib: cdev: support GPIO_V2_GET_LINE_IOCTL and GPIO_V2_LINE_GET_VALUES_IOCTL")
Fixes: aad95584 ("gpiolib: cdev: support GPIO_V2_GET_LINEINFO_IOCTL and GPIO_V2_GET_LINEINFO_WATCH_IOCTL")
Fixes: a54756cb ("gpiolib: cdev: support GPIO_V2_LINE_SET_CONFIG_IOCTL")
Fixes: 7b8e00d9 ("gpiolib: cdev: support GPIO_V2_LINE_SET_VALUES_IOCTL")
Signed-off-by: NBartosz Golaszewski <bartosz.golaszewski@linaro.org>
[Nick: fixed a build failure with CDEV_V1 disabled]
Co-authored-by: NNick Hainke <vincent@systemli.org>
Reviewed-by: NKent Gibson <warthog618@gmail.com>
Reviewed-by: NAndy Shevchenko <andriy.shevchenko@linux.intel.com>
Reviewed-by: NLinus Walleij <linus.walleij@linaro.org>

Provide to_gpio_device() helper which can be utilized in the existing
and future code.

While at it, make sure it becomes no-op at compilation time.
Signed-off-by: NAndy Shevchenko <andriy.shevchenko@linux.intel.com>
Reviewed-by: NLinus Walleij <linus.walleij@linaro.org>
Signed-off-by: NBartosz Golaszewski <bartosz.golaszewski@linaro.org>

Some of the functions are using struct fwnode_handle, some struct device
pointer. In the GPIO library the firmware node of the GPIO device is the
same as GPIO node of the GPIO chip. Due to this fact we may use former
to access properties everywhere in the code.
Signed-off-by: NAndy Shevchenko <andriy.shevchenko@linux.intel.com>
Reviewed-by: NBrian Masney <bmasney@redhat.com>
Tested-by: NMarijn Suijten <marijn.suijten@somainline.org>
[Bartosz: stick to the 80-char limit where it's not hurting readability]
Signed-off-by: NBartosz Golaszewski <bartosz.golaszewski@linaro.org>

Here is a backtrace report about memory leak detected in
gpiochip_setup_dev():

unreferenced object 0xffff88810b406400 (size 512):
  comm "python3", pid 1682, jiffies 4295346908 (age 24.090s)
  backtrace:
    kmalloc_trace
    device_add		device_private_init at drivers/base/core.c:3361
			(inlined by) device_add at drivers/base/core.c:3411
    cdev_device_add
    gpiolib_cdev_register
    gpiochip_setup_dev
    gpiochip_add_data_with_key

gcdev_register() & gcdev_unregister() would call device_add() &
device_del() (no matter CONFIG_GPIO_CDEV is enabled or not) to
register/unregister device.

However, if device_add() succeeds, some resource (like
struct device_private allocated by device_private_init())
is not released by device_del().

Therefore, after device_add() succeeds by gcdev_register(), it
needs to call put_device() to release resource in the error handle
path.

Here we move forward the register of release function, and let it
release every piece of resource by put_device() instead of kfree().

While at it, fix another subtle issue, i.e. when gc->ngpio is equal
to 0, we still call kcalloc() and, in case of further error, kfree()
on the ZERO_PTR pointer, which is not NULL. It's not a bug per se,
but rather waste of the resources and potentially wrong expectation
about contents of the gdev->descs variable.

Fixes: 159f3cd9 ("gpiolib: Defer gpio device setup until after gpiolib initialization")
Signed-off-by: NZeng Heng <zengheng4@huawei.com>
Co-developed-by: NAndy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: NAndy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: NBartosz Golaszewski <bartosz.golaszewski@linaro.org>

Note that this is a RFC patch and not meant to be merged. I looked into
a problem with linux-next-20221110 on the Qualcomm SA8540P automotive
board (sc8280xp) where the UFS host controller would fail to probe due
to repeated probe deferrals when trying to get reset-gpios via
devm_gpiod_get_optional().

of_get_named_gpiod_flags() returns -EPROBE_DEFER, which is caused by
of_gpiochip_match_node_and_xlate() returning 0 since the of_xlate function
pointer is not set for the qcom,sc8280xp-tlmm pinctrl driver. The
pinctrl driver doesn't define one, so of_gpiochip_add() should
automatically setup of_gpio_simple_xlate() on it's behalf. This doesn't
happen since the fwnode member on the struct gpiochip is set to null
when of_gpiochip_add() is called. Let's work around this by ensuring
that it's set if available.

Note that this broke sometime within the last few weeks within
linux-next and I haven't bisected this. I'm posting this in the hopes
that someone may know offhand which patch(es) may have broken this.
Signed-off-by: NBrian Masney <bmasney@redhat.com>
Tested-by: NMarijn Suijten <marijn.suijten@somainline.org>
Tested-by: NKonrad Dybcio <konrad.dybcio@linaro.org>
Tested-by: Steev Klimaszewski <steev@kali.org> #Lenovo Thinkpad X13s
Tested-by: NNeil Armstrong <neil.armstrong@linaro.org>
Signed-off-by: NBartosz Golaszewski <bartosz.golaszewski@linaro.org>

Now that static device properties understand notion of child nodes and
references, let's teach gpiolib to handle them:

- GPIOs are represented as a references to software nodes representing
  gpiochip
- references must have 2 arguments - GPIO number within the chip and
  GPIO flags (GPIO_ACTIVE_LOW/GPIO_ACTIVE_HIGH, etc)
- a new PROPERTY_ENTRY_GPIO() macro is supplied to ensure the above
- name of the software node representing gpiochip must match label of
  the gpiochip, as we use it to locate gpiochip structure at runtime

The following illustrates use of software nodes to describe a "System"
button that is currently specified via use of gpio_keys_platform_data
in arch/mips/alchemy/board-mtx1.c. It follows bindings specified in
Documentation/devicetree/bindings/input/gpio-keys.yaml.

static const struct software_node mxt1_gpiochip2_node = {
	.name = "alchemy-gpio2",
};

static const struct property_entry mtx1_gpio_button_props[] = {
	PROPERTY_ENTRY_U32("linux,code", BTN_0),
	PROPERTY_ENTRY_STRING("label", "System button"),
	PROPERTY_ENTRY_GPIO("gpios", &mxt1_gpiochip2_node, 7, GPIO_ACTIVE_LOW),
	{ }
};

Similarly, arch/arm/mach-tegra/board-paz00.c can be converted to:

static const struct software_node tegra_gpiochip_node = {
	.name = "tegra-gpio",
};

static struct property_entry wifi_rfkill_prop[] __initdata = {
	PROPERTY_ENTRY_STRING("name", "wifi_rfkill"),
	PROPERTY_ENTRY_STRING("type", "wlan"),
	PROPERTY_ENTRY_GPIO("reset-gpios",
			    &tegra_gpiochip_node, 25, GPIO_ACTIVE_HIGH);
	PROPERTY_ENTRY_GPIO("shutdown-gpios",
			    &tegra_gpiochip_node, 85, GPIO_ACTIVE_HIGH);
	{ },
};

static struct platform_device wifi_rfkill_device = {
	.name	= "rfkill_gpio",
	.id	= -1,
};

...

	software_node_register(&tegra_gpiochip_node);
	device_create_managed_software_node(&wifi_rfkill_device.dev,
					    wifi_rfkill_prop, NULL);
Acked-by: NLinus Walleij <linus.walleij@linaro.org>
Reviewed-by: NAndy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: NDmitry Torokhov <dmitry.torokhov@gmail.com>
Signed-off-by: NBartosz Golaszewski <bartosz.golaszewski@linaro.org>

Ensure that all paths to obtain/look up GPIOD from generic
consumer-visible APIs go through the new gpiod_find_and_request()
helper, so that we can easily extend it with support for new firmware
mechanisms.

The only exception is OF-specific [devm_]gpiod_get_from_of_node() API
that is still being used by a couple of drivers and will be removed as
soon as patches converting them to use generic fwnode/device APIs are
accepted.
Acked-by: NLinus Walleij <linus.walleij@linaro.org>
Signed-off-by: NDmitry Torokhov <dmitry.torokhov@gmail.com>
Signed-off-by: NBartosz Golaszewski <bartosz.golaszewski@linaro.org>

There is no need for the generic parts of GPIOLIB to be aware of
implementation details of ACPI-bases lookups.
Reviewed-by: NAndy Shevchenko <andriy.shevchenko@linux.intel.com>
Acked-by: NLinus Walleij <linus.walleij@linaro.org>
Signed-off-by: NDmitry Torokhov <dmitry.torokhov@gmail.com>
Signed-off-by: NBartosz Golaszewski <bartosz.golaszewski@linaro.org>

In preparation of switching all ACPI-based GPIO lookups to go through
acpi_find_gpio() let's change it to accept device node as its argument
as we do not always have access to device structure.
Reviewed-by: NAndy Shevchenko <andriy.shevchenko@linux.intel.com>
Acked-by: NLinus Walleij <linus.walleij@linaro.org>
Signed-off-by: NDmitry Torokhov <dmitry.torokhov@gmail.com>
Signed-off-by: NBartosz Golaszewski <bartosz.golaszewski@linaro.org>

In preparation of switching all OF-based GPIO lookups to go through
of_find_gpio() let's change it to accept device node as its argument as
we do not always have access to device structure.
Reviewed-by: NAndy Shevchenko <andriy.shevchenko@linux.intel.com>
Acked-by: NLinus Walleij <linus.walleij@linaro.org>
Signed-off-by: NDmitry Torokhov <dmitry.torokhov@gmail.com>
Signed-off-by: NBartosz Golaszewski <bartosz.golaszewski@linaro.org>

In preparation to complete fwnode switch, integrate
of_gpiochip_init_valid_mask() into gpiochip_init_valid_mask().
Signed-off-by: NAndy Shevchenko <andriy.shevchenko@linux.intel.com>
Reviewed-by: NDmitry Torokhov <dmitry.torokhov@gmail.com>
Reviewed-by: NLinus Walleij <linus.walleij@linaro.org>
Signed-off-by: NBartosz Golaszewski <bartosz.golaszewski@linaro.org>

Since commit 14e85c0e ("gpio: remove gpio_descs global array")
there is no limitation on the number of GPIOs that can be allocated
in the system since the allocation is fully dynamic.

ARCH_NR_GPIOS is today only used in order to provide downwards
gpiobase allocation from that value, while static allocation is
performed upwards from 0. However that has the disadvantage of
limiting the number of GPIOs that can be registered in the system.

To overcome this limitation without requiring each and every
platform to provide its 'best-guess' maximum number, rework the
allocation to allocate upwards, allowing approx 2 millions of
GPIOs.

In order to still allow static allocation for legacy drivers, define
GPIO_DYNAMIC_BASE with the value 512 as the start for dynamic
allocation. The 512 value is chosen because it is the end of
the current default range so all current static allocations are
expected to be below that value. Of course that's just a rough
estimate based on the default value, but assuming static
allocations come first, even if there are more static allocations
it should fit under the 512 value.

In the future, it is expected that all static allocations go away
and then dynamic allocation will be patched to start at 0.
Signed-off-by: NChristophe Leroy <christophe.leroy@csgroup.eu>
Reviewed-by: NAndy Shevchenko <andy.shevchenko@gmail.com>
Signed-off-by: NBartosz Golaszewski <bartosz.golaszewski@linaro.org>

In the preparation of getting completely rid of static gpiobase
allocation in the future, emit a warning in drivers still doing so.
Signed-off-by: NChristophe Leroy <christophe.leroy@csgroup.eu>
Reviewed-by: NAndy Shevchenko <andy.shevchenko@gmail.com>
Signed-off-by: NBartosz Golaszewski <bartosz.golaszewski@linaro.org>

There are no external users of fwnode_get_named_gpiod() anymore, so
let's stop exporting it and mark it as static.
Signed-off-by: NDmitry Torokhov <dmitry.torokhov@gmail.com>
Reviewed-by: NAndy Shevchenko <andy.shevchenko@gmail.com>
Signed-off-by: NBartosz Golaszewski <brgl@bgdev.pl>

This change prepares the gpio core to look at firmware flags and set
'FLAG_BIAS_DISABLE' if necessary. It works in similar way to
'GPIO_PULL_DOWN' and 'GPIO_PULL_UP'.
Signed-off-by: NNuno Sá <nuno.sa@analog.com>
Signed-off-by: NBartosz Golaszewski <brgl@bgdev.pl>

Allow free() callback to be overridden from irq_domain_ops for
hierarchical chips.

This allows drivers to free up resources which are allocated during
child_to_parent_hwirq()/populate_parent_alloc_arg() callbacks.

On Renesas RZ/G2L platform a bitmap is maintained for TINT slots, a slot
is allocated in child_to_parent_hwirq() callback which is freed up in free
callback hence this override.
Signed-off-by: NLad Prabhakar <prabhakar.mahadev-lad.rj@bp.renesas.com>
Reviewed-by: NGeert Uytterhoeven <geert+renesas@glider.be>
Acked-by: NBartosz Golaszewski <brgl@bgdev.pl>
Signed-off-by: NMarc Zyngier <maz@kernel.org>
Link: https://lore.kernel.org/r/20220707182314.66610-5-prabhakar.mahadev-lad.rj@bp.renesas.com

The gpiolib is unique in the way it uses intermediate fwspecs
when feeding an interrupt specifier to the parent domain, as it
relies on the populate_parent_alloc_arg() callback to perform
a dynamic allocation.

This is pretty inefficient (we free the structure almost immediately),
and the only reason this isn't a stack allocation is that our
ThunderX friend uses MSIs rather than a FW-constructed structure.

Let's solve it by providing a new type composed of the union
of a struct irq_fwspec and a msi_info_t, which satisfies both
requirements. This allows us to use a stack allocation, and we
can move the handful of users to this new scheme.

Also perform some additional cleanup, such as getting rid of the
stub versions of the irq_domain_translate_*cell helpers, which
are never used when CONFIG_IRQ_DOMAIN_HIERARCHY isn't selected.

Tested on a Tegra186.
Reviewed-by: NLinus Walleij <linus.walleij@linaro.org>
Signed-off-by: NMarc Zyngier <maz@kernel.org>
Cc: Daniel Palmer <daniel@thingy.jp>
Cc: Romain Perier <romain.perier@gmail.com>
Cc: Bartosz Golaszewski <brgl@bgdev.pl>
Cc: Thierry Reding <thierry.reding@gmail.com>
Cc: Jonathan Hunter <jonathanh@nvidia.com>
Cc: Robert Richter <rric@kernel.org>
Cc: Nobuhiro Iwamatsu <nobuhiro1.iwamatsu@toshiba.co.jp>
Cc: Andy Gross <agross@kernel.org>
Cc: Bjorn Andersson <bjorn.andersson@linaro.org>
Acked-by: NBartosz Golaszewski <brgl@bgdev.pl>
Link: https://lore.kernel.org/r/20220707182314.66610-2-prabhakar.mahadev-lad.rj@bp.renesas.com

Some GPIO chip can provide hardware timestamp support on its GPIO lines
, in order to support that, additional API needs to be added which
can talk to both GPIO chip and HTE (hardware timestamping engine)
providers if there is any dependencies. This patch introduces optional
hooks to enable and disable hardware timestamping related features
in the GPIO controller chip.
Signed-off-by: NDipen Patel <dipenp@nvidia.com>
Reviewed-by: NLinus Walleij <linus.walleij@linaro.org>
Reported-by: Nkernel test robot <lkp@intel.com>
Signed-off-by: NThierry Reding <treding@nvidia.com>

Commit 5467801f ("gpio: Restrict usage of GPIO chip irq members
before initialization") attempted to fix a race condition that lead to a
NULL pointer, but in the process caused a regression for _AEI/_EVT
declared GPIOs.

This manifests in messages showing deferred probing while trying to
allocate IRQs like so:

  amd_gpio AMDI0030:00: Failed to translate GPIO pin 0x0000 to IRQ, err -517
  amd_gpio AMDI0030:00: Failed to translate GPIO pin 0x002C to IRQ, err -517
  amd_gpio AMDI0030:00: Failed to translate GPIO pin 0x003D to IRQ, err -517
  [ .. more of the same .. ]

The code for walking _AEI doesn't handle deferred probing and so this
leads to non-functional GPIO interrupts.

Fix this issue by moving the call to `acpi_gpiochip_request_interrupts`
to occur after gc->irc.initialized is set.

Fixes: 5467801f ("gpio: Restrict usage of GPIO chip irq members before initialization")
Link: https://lore.kernel.org/linux-gpio/BL1PR12MB51577A77F000A008AA694675E2EF9@BL1PR12MB5157.namprd12.prod.outlook.com/
Link: https://bugzilla.suse.com/show_bug.cgi?id=1198697
Link: https://bugzilla.kernel.org/show_bug.cgi?id=215850
Link: https://gitlab.freedesktop.org/drm/amd/-/issues/1979
Link: https://gitlab.freedesktop.org/drm/amd/-/issues/1976Reported-by: NMario Limonciello <mario.limonciello@amd.com>
Signed-off-by: NMario Limonciello <mario.limonciello@amd.com>
Reviewed-by: NShreeya Patel <shreeya.patel@collabora.com>
Tested-By: NSamuel Čavoj <samuel@cavoj.net>
Tested-By: lukeluk498@gmail.com Link:
Reviewed-by: NAndy Shevchenko <andy.shevchenko@gmail.com>
Acked-by: NLinus Walleij <linus.walleij@linaro.org>
Reviewed-and-tested-by: NTakashi Iwai <tiwai@suse.de>
Cc: Shreeya Patel <shreeya.patel@collabora.com>
Cc: stable@vger.kernel.org
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

The GPIO subsystem has a couple of internal helpers to manage
resources on behalf of the irqchip. Expose them so that GPIO
drivers can use them directly.
Reviewed-by: NAndy Shevchenko <andy.shevchenko@gmail.com>
Reviewed-by: NBartosz Golaszewski <brgl@bgdev.pl>
Signed-off-by: NMarc Zyngier <maz@kernel.org>
Link: https://lore.kernel.org/r/20220419141846.598305-3-maz@kernel.org

In order to move away from gpiolib messing with the internals of
unsuspecting irqchips, add a flag by which irqchips advertise
that they are not to be messed with, and do solemnly swear that
they correctly call into the gpiolib helpers when required.

Also nudge the users into converting their drivers to the
new model.
Reviewed-by: NAndy Shevchenko <andy.shevchenko@gmail.com>
Reviewed-by: NBartosz Golaszewski <brgl@bgdev.pl>
Signed-off-by: NMarc Zyngier <maz@kernel.org>
Link: https://lore.kernel.org/r/20220419141846.598305-2-maz@kernel.org

In couple of cases we are using the same code to wrap ->get() callback.
Extract that code into a helper for the sake of better maintenance.
Signed-off-by: NAndy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: NBartosz Golaszewski <brgl@bgdev.pl>

Use for_each_gpio_desc() and since we would need to touch the entire
conditionals, do the following:
- rename last occurrence of gdesc to desc
- use short ternary operator ?:
- join two seq_printf() calls into single one
- fix indentation of the seq_printf() parameters
Signed-off-by: NAndy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: NBartosz Golaszewski <brgl@bgdev.pl>

In some cases we want to traverse all GPIO descriptors for given
chip, let's split out for_each_gpio_desc() macro for such cases.
Signed-off-by: NAndy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: NBartosz Golaszewski <brgl@bgdev.pl>

The iterator loop is used exclusively to get a descriptor, which in its
turn is what is being used by the caller. Embed the iterator variable
into the loop in the for_each_gpio_desc_with_flag() macro helper.
Suggested-by: NBartosz Golaszewski <brgl@bgdev.pl>
Signed-off-by: NAndy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: NBartosz Golaszewski <brgl@bgdev.pl>

An error path is a slow path, no need to block other CPUs
when printing error messages.
Signed-off-by: NAndy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: NBartosz Golaszewski <brgl@bgdev.pl>

In the snippets like the following

	if (...)
		return / goto / break / continue ...;
	else
		...

the 'else' is redundant. Get rid of it. In case of IOCTLs use
switch-case pattern that seems the usual in such cases.

While at it, clarify necessity of else in gpiod_direction_output()
by attaching else if to the closing curly brace on a previous line.
Signed-off-by: NAndy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: NBartosz Golaszewski <brgl@bgdev.pl>

GPIO chip irq members are exposed before they could be completely
initialized and this leads to race conditions.

One such issue was observed for the gc->irq.domain variable which
was accessed through the I2C interface in gpiochip_to_irq() before
it could be initialized by gpiochip_add_irqchip(). This resulted in
Kernel NULL pointer dereference.

Following are the logs for reference :-

kernel: Call Trace:
kernel:  gpiod_to_irq+0x53/0x70
kernel:  acpi_dev_gpio_irq_get_by+0x113/0x1f0
kernel:  i2c_acpi_get_irq+0xc0/0xd0
kernel:  i2c_device_probe+0x28a/0x2a0
kernel:  really_probe+0xf2/0x460
kernel: RIP: 0010:gpiochip_to_irq+0x47/0xc0

To avoid such scenarios, restrict usage of GPIO chip irq members before
they are completely initialized.
Signed-off-by: NShreeya Patel <shreeya.patel@collabora.com>
Cc: stable@vger.kernel.org
Reviewed-by: NAndy Shevchenko <andy.shevchenko@gmail.com>
Reviewed-by: NLinus Walleij <linus.walleij@linaro.org>
Signed-off-by: NBartosz Golaszewski <brgl@bgdev.pl>

This reverts commit fc328a7d.

This commit - while attempting to fix a regression - has caused a number
of other problems. As the fallout from it is more significant than the
initial problem itself, revert it for now before we find a correct
solution.

Link: https://lore.kernel.org/all/20220314192522.GA3031157@roeck-us.net/
Link: https://lore.kernel.org/stable/20220314155509.552218-1-michael@walle.cc/
Link: https://lore.kernel.org/all/20211217153555.9413-1-marcelo.jimenez@gmail.com/Signed-off-by: NBartosz Golaszewski <brgl@bgdev.pl>
Reported-and-bisected-by: NGuenter Roeck <linux@roeck-us.net>
Reported-by: NMichael Walle <michael@walle.cc>
Cc: Thorsten Leemhuis <linux@leemhuis.info>
Cc: Marcelo Roberto Jimenez <marcelo.jimenez@gmail.com>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Use list_first_entry()/list_last_entry() instead of open coded variants.
Signed-off-by: NAndy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: NBartosz Golaszewski <brgl@bgdev.pl>

It appears that GPIO ACPI library uses ACPI debounce values directly.
However, the GPIO library APIs expect the debounce timeout to be in
microseconds.

Convert ACPI value of debounce to microseconds.

While at it, document this detail where it is appropriate.

BugLink: https://bugzilla.kernel.org/show_bug.cgi?id=215664Reported-by: NKai-Heng Feng <kai.heng.feng@canonical.com>
Fixes: 8dcb7a15 ("gpiolib: acpi: Take into account debounce settings")
Signed-off-by: NAndy Shevchenko <andriy.shevchenko@linux.intel.com>
Tested-by: NKai-Heng Feng <kai.heng.feng@canonical.com>
Reviewed-by: NMika Westerberg <mika.westerberg@linux.intel.com>
Signed-off-by: NBartosz Golaszewski <brgl@bgdev.pl>

Some GPIO lines have stopped working after the patch
commit 2ab73c6d ("gpio: Support GPIO controllers without pin-ranges")

And this has supposedly been fixed in the following patches
commit 89ad556b ("gpio: Avoid using pin ranges with !PINCTRL")
commit 6dbbf846 ("gpiolib: Don't free if pin ranges are not defined")

But an erratic behavior where some GPIO lines work while others do not work
has been introduced.

This patch reverts those changes so that the sysfs-gpio interface works
properly again.
Signed-off-by: NMarcelo Roberto Jimenez <marcelo.jimenez@gmail.com>
Signed-off-by: NBartosz Golaszewski <brgl@bgdev.pl>

We are racing the registering of .to_irq when probing the
i2c driver. This results in random failure of touchscreen
devices.

Following explains the race condition better.

[gpio driver] gpio driver registers gpio chip
[gpio consumer] gpio is acquired
[gpio consumer] gpiod_to_irq() fails with -ENXIO
[gpio driver] gpio driver registers irqchip
gpiod_to_irq works at this point, but -ENXIO is fatal

We could see the following errors in dmesg logs when gc->to_irq is NULL

[2.101857] i2c_hid i2c-FTS3528:00: HID over i2c has not been provided an Int IRQ
[2.101953] i2c_hid: probe of i2c-FTS3528:00 failed with error -22

To avoid this situation, defer probing until to_irq is registered.
Returning -EPROBE_DEFER would be the first step towards avoiding
the failure of devices due to the race in registration of .to_irq.
Final solution to this issue would be to avoid using gc irq members
until they are fully initialized.

This issue has been reported many times in past and people have been
using workarounds like changing the pinctrl_amd to built-in instead
of loading it as a module or by adding a softdep for pinctrl_amd into
the config file.

BugLink: https://bugzilla.kernel.org/show_bug.cgi?id=209413Reviewed-by: NLinus Walleij <linus.walleij@linaro.org>
Reviewed-by: NAndy Shevchenko <andy.shevchenko@gmail.com>
Reported-by: Nkernel test robot <lkp@intel.com>
Signed-off-by: NShreeya Patel <shreeya.patel@collabora.com>
Signed-off-by: NBartosz Golaszewski <brgl@bgdev.pl>

Simplify error path in the gpiod_get_index() when requesting a GPIO line by:
 - checking for error condition first
 - dropping redundant 'else'

As a result, decrease the indentation level for better readability.
Signed-off-by: NAndy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: NBartosz Golaszewski <brgl@bgdev.pl>

Instead of repeating first argument for true branch, use short
form of the ternary operator, i.e. ?:.

While at it, fix a typo in the comment.
Signed-off-by: NAndy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: NBartosz Golaszewski <brgl@bgdev.pl>

In a few places we are using a loop against all GPIO descriptors
with a given flag for a given device. Replace it with a consolidated
for_each type of macro.
Signed-off-by: NAndy Shevchenko <andriy.shevchenko@linux.intel.com>
Reviewed-by: NLinus Walleij <linus.walleij@linaro.org>
Signed-off-by: NBartosz Golaszewski <brgl@bgdev.pl>

Some GPIO providers set names for GPIO lines that match the names of
the pins on the SoC, or variations on that theme. These names are
generic more often that not, such as pioC12 in the at91 case. These
generic names block the possibility to set more useful GPIO line
names with device properties (i.e. gpio-line-names).

Allow overriding a generic name given by the GPIO driver if there is
a name given to the GPIO line using device properties, but leave the
generic name alone if no better name is available.

However, there is a risk. If user space is depending on the above
mentioned fixed GPIO names, AND there are device properties that
previously did not reach the surface, the name change might cause
regressions. But hopefully this stays below the radar...
Reviewed-by: NAndy Shevchenko <andy.shevchenko@gmail.com>
Tested-by: NAlexander Dahl <ada@thorsis.com>
Signed-off-by: NPeter Rosin <peda@axentia.se>
Signed-off-by: NBartosz Golaszewski <brgl@bgdev.pl>

Software nodes allow us to represent hierarchies for device components
that don't have their struct device representation yet - for instance:
banks of GPIOs under a common GPIO expander. The core gpiolib core
however doesn't offer any way of passing this information from the
drivers.

This extends struct gpio_chip with a pointer to fwnode that can be set
by the driver and used to pass device properties for child nodes.

This is similar to how we handle device-tree sub-nodes with
CONFIG_OF_GPIO enabled.
Signed-off-by: NBartosz Golaszewski <brgl@bgdev.pl>
Reviewed-by: NAndy Shevchenko <andriy.shevchenko@linux.intel.com>
Acked-by: NLinus Walleij <linus.walleij@linaro.org>

Currently all users of gpiod_add_hogs() call it only once at system
init so there never was any need for a mechanism allowing to remove
them. Now the upcoming gpio-sim will need to tear down chips with hogged
lines so provide a function that allows to remove hogs.
Signed-off-by: NBartosz Golaszewski <brgl@bgdev.pl>
Reviewed-by: NAndy Shevchenko <andriy.shevchenko@linux.intel.com>
Acked-by: NLinus Walleij <linus.walleij@linaro.org>