linux-stable.git/drivers/gpio, branch v6.6.9

gpio: dwapb: mask/unmask IRQ when disable/enale it

2024-01-01T12:42:46+00:00

commit 1cc3542c76acb5f59001e3e562eba672f1983355 upstream.

In the hardware implementation of the I2C HID driver based on DesignWare
GPIO IRQ chip, when the user continues to use the I2C HID device in the
suspend process, the I2C HID interrupt will be masked after the resume
process is finished.

This is because the disable_irq()/enable_irq() of the DesignWare GPIO
driver does not synchronize the IRQ mask register state. In normal use
of the I2C HID procedure, the GPIO IRQ irq_mask()/irq_unmask() functions
are called in pairs. In case of an exception, i2c_hid_core_suspend()
calls disable_irq() to disable the GPIO IRQ. With low probability, this
causes irq_unmask() to not be called, which causes the GPIO IRQ to be
masked and not unmasked in enable_irq(), raising an exception.

Add synchronization to the masked register state in the
dwapb_irq_enable()/dwapb_irq_disable() function. mask the GPIO IRQ
before disabling it. After enabling the GPIO IRQ, unmask the IRQ.

Fixes: 7779b3455697 ("gpio: add a driver for the Synopsys DesignWare APB GPIO block")
Cc: stable@kernel.org
Co-developed-by: Riwen Lu 
Signed-off-by: Riwen Lu 
Signed-off-by: xiongxin 
Acked-by: Serge Semin 
Reviewed-by: Andy Shevchenko 
Signed-off-by: Bartosz Golaszewski 
Signed-off-by: Greg Kroah-Hartman

gpiolib: cdev: add gpio_device locking wrapper around gpio_ioctl()

2024-01-01T12:42:35+00:00

[ Upstream commit 1d656bd259edb89dc1d9938ec5c5389867088546 ]

While the GPIO cdev gpio_ioctl() call is in progress, the kernel can
call gpiochip_remove() which will set gdev->chip to NULL, after which
any subsequent access will cause a crash.

gpio_ioctl() was overlooked by the previous fix to protect syscalls
(bdbbae241a04), so add protection for that.

Fixes: bdbbae241a04 ("gpiolib: protect the GPIO device against being dropped while in use by user-space")
Fixes: d7c51b47ac11 ("gpio: userspace ABI for reading/writing GPIO lines")
Fixes: 3c0d9c635ae2 ("gpiolib: cdev: support GPIO_V2_GET_LINE_IOCTL and GPIO_V2_LINE_GET_VALUES_IOCTL")
Fixes: aad955842d1c ("gpiolib: cdev: support GPIO_V2_GET_LINEINFO_IOCTL and GPIO_V2_GET_LINEINFO_WATCH_IOCTL")
Signed-off-by: Kent Gibson 
Signed-off-by: Bartosz Golaszewski 
Signed-off-by: Sasha Levin

gpiolib: sysfs: Fix error handling on failed export

2023-12-13T17:45:30+00:00

[ Upstream commit 95dd1e34ff5bbee93a28ff3947eceaf6de811b1a ]

If gpio_set_transitory() fails, we should free the GPIO again. Most
notably, the flag FLAG_REQUESTED has previously been set in
gpiod_request_commit(), and should be reset on failure.

To my knowledge, this does not affect any current users, since the
gpio_set_transitory() mainly returns 0 and -ENOTSUPP, which is converted
to 0. However the gpio_set_transitory() function calles the .set_config()
function of the corresponding GPIO chip and there are some GPIO drivers in
which some (unlikely) branches return other values like -EPROBE_DEFER,
and -EINVAL. In these cases, the above mentioned FLAG_REQUESTED would not
be reset, which results in the pin being blocked until the next reboot.

Fixes: e10f72bf4b3e ("gpio: gpiolib: Generalise state persistence beyond sleep")
Signed-off-by: Boerge Struempfel 
Reviewed-by: Andy Shevchenko 
Signed-off-by: Bartosz Golaszewski 
Signed-off-by: Sasha Levin

gpiolib: of: Add quirk for mt2701-cs42448 ASoC sound

2023-11-28T17:19:43+00:00

[ Upstream commit 9e189e80dcb68528dea9e061d9704993f98cb84f ]

These gpio names are due to old DT bindings not following the
"-gpio"/"-gpios" conventions. Handle it using a quirk so the
driver can just look up the GPIOs.

Signed-off-by: Linus Walleij 
Acked-by: Bartosz Golaszewski 
Reviewed-by: AngeloGioacchino Del Regno 
Link: https://lore.kernel.org/r/20231006-descriptors-asoc-mediatek-v1-1-07fe79f337f5@linaro.org
Signed-off-by: Mark Brown 
Signed-off-by: Sasha Levin

gpiolib: acpi: Add a ignore interrupt quirk for Peaq C1010

2023-11-28T17:19:38+00:00

[ Upstream commit 6cc64f6173751d212c9833bde39e856b4f585a3e ]

On the Peaq C1010 2-in-1 INT33FC:00 pin 3 is connected to
a "dolby" button. At the ACPI level an _AEI event-handler
is connected which sets an ACPI variable to 1 on both
edges. This variable can be polled + cleared to 0 using WMI.

Since the variable is set on both edges the WMI interface is pretty
useless even when polling. So instead of writing a custom WMI
driver for this the x86-android-tablets code instantiates
a gpio-keys platform device for the "dolby" button.

Add an ignore_interrupt quirk for INT33FC:00 pin 3 on the Peaq C1010,
so that it is not seen as busy when the gpio-keys driver requests it.

Note this replaces a hack in x86-android-tablets where it would
call acpi_gpiochip_free_interrupts() on the INT33FC:00 GPIO
controller. acpi_gpiochip_free_interrupts() is considered private
(internal) gpiolib API so x86-android-tablets should stop using it.

Signed-off-by: Hans de Goede 
Reviewed-by: Andy Shevchenko 
Reviewed-by: Mika Westerberg 
Acked-by: Linus Walleij 
Acked-by: Bartosz Golaszewski 
Link: https://lore.kernel.org/r/20230909141816.58358-3-hdegoede@redhat.com
Signed-off-by: Sasha Levin

gpio: sim: initialize a managed pointer when declaring it

2023-11-20T10:59:04+00:00

[ Upstream commit 9f93f18305f5777820491e6ab9b34422c160371b ]

Variables managed with __free() should typically be initialized where
they are declared so that the __free() callback is paired with its
counterpart resource allocator. Fix the second instance of using
__free() in gpio-sim to follow this pattern.

Fixes: 3faf89f27aab ("gpio: sim: simplify code with cleanup helpers")
Suggested-by: Linus Torvalds 
Signed-off-by: Bartosz Golaszewski 
Signed-off-by: Sasha Levin

gpiolib: acpi: Add missing memset(0) to acpi_get_gpiod_from_data()

2023-10-20T09:24:16+00:00

When refactoring the acpi_get_gpiod_from_data() the change missed
cleaning up the variable on stack. Add missing memset().

Reported-by: Ferry Toth 
Fixes: 16ba046e86e9 ("gpiolib: acpi: teach acpi_find_gpio() to handle data-only nodes")
Signed-off-by: Andy Shevchenko 
Reviewed-by: Dmitry Torokhov 
Signed-off-by: Bartosz Golaszewski

gpio: vf610: set value before the direction to avoid a glitch

2023-10-18T09:01:08+00:00

We found a glitch when configuring the pad as output high. To avoid this
glitch, move the data value setting before direction config in the
function vf610_gpio_direction_output().

Fixes: 659d8a62311f ("gpio: vf610: add imx7ulp support")
Signed-off-by: Haibo Chen 
[Bartosz: tweak the commit message]
Signed-off-by: Bartosz Golaszewski

gpio: vf610: mask the gpio irq in system suspend and support wakeup

2023-10-18T08:49:04+00:00

Add flag IRQCHIP_MASK_ON_SUSPEND to make sure gpio irq is masked on
suspend, if lack this flag, current irq arctitecture will not mask
the irq, and these unmasked gpio irq will wrongly wakeup the system
even they are not config as wakeup source.

Also add flag IRQCHIP_ENABLE_WAKEUP_ON_SUSPEND to make sure the gpio
irq which is configed as wakeup source can work as expect.

Fixes: 7f2691a19627 ("gpio: vf610: add gpiolib/IRQ chip driver for Vybrid")
Signed-off-by: Haibo Chen 
Signed-off-by: Bartosz Golaszewski

gpio: aspeed: fix the GPIO number passed to pinctrl_gpio_set_config()

2023-10-05T06:41:56+00:00

pinctrl_gpio_set_config() expects the GPIO number from the global GPIO
numberspace, not the controller-relative offset, which needs to be added
to the chip base.

Fixes: 5ae4cb94b313 ("gpio: aspeed: Add debounce support")
Signed-off-by: Bartosz Golaszewski 
Reviewed-by: Andy Shevchenko 
Reviewed-by: Andrew Jeffery