kernel/linux.git/drivers/gpio/gpiolib.c, branch v7.1

gpiolib: handle gpio-hogs only once

2026-06-10T08:12:57+00:00

Commit d1d564ec49929 ("gpio: move hogs into GPIO core") introduced a behaviour change that breaks boot on Raspberry Pi 5 when using the firmware-supplied device tree: gpiochip_add_data_with_key: GPIOs 544..575 (/soc@107c000000/gpio@7d517c00) failed to register, -22 brcmstb-gpio 107d517c00.gpio: Could not add gpiochip for bank 1 brcmstb-gpio 107d517c00.gpio: probe with driver brcmstb-gpio failed with error -22 gpio-brcmstb registers two gpio_chips against the device tree node gpio@7d517c00, one for each bank. The firmware-supplied DT includes a gpio-hog on RP1 RUN, and this gpio-hog is attempted to be applied to *both* gpio_chips. This succeeds against bank 0 (which hosts the GPIO) and fails for bank 1 (which does not). In the previous implementation, failures to apply gpio-hogs were quietly ignored. In the new code, the error code propagates and causes probe to fail. Closely approximate the previous behaviour by using the OF_POPULATED flag to ensure that each gpio-hog is processed only once. The flag was previously being set before the gpio-hogs were processed, so as part of this change, the flag now gets set only after the gpio-hog is actioned. The handling of gpio-hogs on a DT node with multiple gpio_chips remains a bit incomplete/unclear, but this at least retains the ability to apply hogs to the first gpio_chip per node. Fixes: d1d564ec49929 ("gpio: move hogs into GPIO core") Signed-off-by: Daniel Drake Link: https://patch.msgid.link/20260608210108.36248-1-dan@reactivated.net Signed-off-by: Bartosz Golaszewski

gpio: fix cleanup path on hog failure

2026-06-10T07:29:19+00:00

If gpiochip_hog_lines() successfully processes some hogs but fails on a later one, the error handling path in gpiochip_add_data_with_key() jumps directly to err_remove_of_chip. This leaks resources allocated earlier for ACPI, interrupts and hogs that were successfully processed. Use the right label in error path. Closes: https://sashiko.dev/#/patchset/20260608210108.36248-1-dan%40reactivated.net Fixes: d1d564ec4992 ("gpio: move hogs into GPIO core") Reviewed-by: Andy Shevchenko Reviewed-by: Mika Westerberg Link: https://patch.msgid.link/20260609-gpio-hogs-fixes-v1-2-b4064f8070e7@oss.qualcomm.com Signed-off-by: Bartosz Golaszewski

Merge tag 'gpio-updates-for-v7.1-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/brgl/linux

2026-04-14T03:10:58+00:00

Pull gpio updates from Bartosz Golaszewski: "For this merge window we have two new drivers: support for GPIO-signalled ACPI events on Intel platforms and a generic GPIO-over-pinctrl driver using the ARM SCMI protocol for controlling pins. Several things have been reworked in GPIO core: we unduplicated GPIO hog handling, reduced the number of SRCU locks and dereferences, improved support for software-node-based lookup and removed more legacy code after converting remaining users to modern alternatives. There's also a number of driver reworks and refactoring, documentation updates, some bug-fixes and new tests. GPIO core: - defer probe on software node lookups when the remote software node exists but has not been registered as a firmware node yet - unify GPIO hog handling by moving code duplicated in OF and ACPI modules into GPIO core and allow setting up hogs with software nodes - allow matching GPIO controllers by secondary firmware node if matching by primary does not succeed - demote deferral warnings to debug level as they are quite normal when using software nodes which don't support fw_devlink yet - disable the legacy GPIO character device uAPI v1 supprt in Kconfig by default - rework several core functions in preparation for the upcoming Revocable helper library for protecting resources against sudden removal, this reduces the number of SRCU dereferences in GPIO core - simplify file descriptor logic in GPIO character device code by using FD_PREPARE() - introduce a header defining symbols used by both GPIO consumers and providers to avoid having to include provider-specific headers from drivers which only consume GPIOs - replace snprintf() with strscpy() where formatting is not required New drivers: - add the gpio-by-pinctrl generic driver using the ARM SCMI protocol to control GPIOs (along with SCMI changes pulled from the pinctrl tree) - add a driver providing support for handling of platform events via GPIO-signalled ACPI events (used on Intel Nova Lake and later platforms) Driver changes: - extend the gpio-kempld driver with support for more recent models, interrupts and setting/getting multiple values at once - improve interrupt handling in gpio-brcmstb - add support for multi-SoC systems in gpio-tegra186 - make sure we return correct values from the .get() callbacks in several GPIO drivers by normalizing any values other than 0, 1 or negative error numbers - use flexible arrays in several drivers to reduce the number of required memory allocations - simplify synchronous waiting for virtual drivers to probe and remove the dedicated, a bit overengineered helper library dev-sync-probe - remove unneeded Kconfig dependencies on OF_GPIO in several drivers and subsystems - convert the two remaining users of of_get_named_gpio() to using GPIO descriptors and remove the (no longer used) function along with the header that declares it - add missing includes in gpio-mmio - shrink and simplify code in gpio-max732x by using guard(mutex) - remove duplicated code handling the 'ngpios' property from gpio-ts4800, it's already handled in GPIO core - use correct variable type in gpio-aspeed - add support for a new model in gpio-realtek-otto - allow to specify the active-low setting of simulated hogs over the configfs interface (in addition to existing devicetree support) in gpio-sim Bug fixes: - clear the OF_POPULATED flag on hog nodes in GPIO chip remove path on OF systems - fix resource leaks in error path in gpiochip_add_data_with_key() - drop redundant device reference in gpio-mpsse Tests: - add selftests for use-after-free cases in GPIO character device code DT bindings: - add a DT binding document for SCMI based, gpio-over-pinctrl devices - fix interrupt description in microchip,mpfs-gpio - add new compatible for gpio-realtek-otto - describe the resets of the mpfs-gpio controller - fix maintainer's email in gpio-delay bindings - remove the binding document for cavium,thunder-8890 as the corresponding device is bound over PCI and not firmware nodes Documentation: - update the recommended way of converting legacy boards to using software nodes for GPIO description - describe GPIO line value semantics - misc updates to kerneldocs Misc: - convert OMAP1 ams-delta board to using GPIO hogs described with software nodes" * tag 'gpio-updates-for-v7.1-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/brgl/linux: (79 commits) gpio: swnode: defer probe on references to unregistered software nodes dt-bindings: gpio: cavium,thunder-8890: Remove DT binding Documentation: gpio: update the preferred method for using software node lookup gpio: gpio-by-pinctrl: s/used to do/is used to do/ gpio: aspeed: fix unsigned long int declaration gpio: rockchip: convert to dynamic GPIO base allocation gpio: remove dev-sync-probe gpio: virtuser: stop using dev-sync-probe gpio: aggregator: stop using dev-sync-probe gpio: sim: stop using dev-sync-probe gpio: Add Intel Nova Lake ACPI GPIO events driver gpiolib: Make deferral warnings debug messages gpiolib: fix hogs with multiple lines gpio: fix up CONFIG_OF dependencies gpio: gpio-by-pinctrl: add pinctrl based generic GPIO driver gpio: dt-bindings: Add GPIO on top of generic pin control firmware: arm_scmi: Allow PINCTRL_REQUEST to return EOPNOTSUPP pinctrl: scmi: ignore PIN_CONFIG_PERSIST_STATE pinctrl: scmi: Delete PIN_CONFIG_OUTPUT_IMPEDANCE_OHMS support pinctrl: scmi: Add SCMI_PIN_INPUT_VALUE ...

gpiolib: Make deferral warnings debug messages

2026-04-03T09:09:38+00:00

With the recent addition of the shared GPIO support, warning messages such as the following are being observed ... reg-fixed-voltage regulator-vdd-3v3-pcie: cannot find GPIO chip gpiolib_shared.proxy.6, deferring These are seen even with GPIO_SHARED_PROXY=y. Given that the GPIOs are successfully found a bit later during boot and the code is intentionally returning -EPROBE_DEFER when they are not found, downgrade these messages to debug prints to avoid unnecessary warnings being observed. Note that although the 'cannot find GPIO line' warning has not been observed in this case, it seems reasonable to make this print a debug print for consistency too. Signed-off-by: Jon Hunter Link: https://patch.msgid.link/20260401133441.47641-1-jonathanh@nvidia.com Signed-off-by: Bartosz Golaszewski

gpiolib: fix hogs with multiple lines

2026-03-30T13:14:58+00:00

After moving GPIO hog handling into GPIOLIB core, we accidentally stopped supporting devicetree hog definitions with multiple lines like so: hog { gpio-hog; gpios = <3 0>, <4 GPIO_ACTIVE_LOW>; output-high; line-name = "foo"; }; Restore this functionality to fix reported regressions. Fixes: d1d564ec4992 ("gpio: move hogs into GPIO core") Reported-by: Geert Uytterhoeven Closes: https://lore.kernel.org/all/CAMuHMdX6RuZXAozrF5m625ZepJTVVr4pcyKczSk12MedWvoejw@mail.gmail.com/ Tested-by: Geert Uytterhoeven Reviewed-by: Geert Uytterhoeven Link: https://patch.msgid.link/20260330-gpio-hogs-multiple-v3-1-175c3839ad9f@oss.qualcomm.com Signed-off-by: Bartosz Golaszewski

gpio: shared: handle pins shared by child nodes of devices

2026-03-23T08:54:24+00:00

Shared GPIOs may be assigned to child nodes of device nodes which don't themselves bind to any struct device. We need to pass the firmware node that is the actual consumer to gpiolib-shared and compare against it instead of unconditionally using the fwnode of the consumer device. Fixes: a060b8c511ab ("gpiolib: implement low-level, shared GPIO support") Reported-by: Jon Hunter Closes: https://lore.kernel.org/all/921ba8ce-b18e-4a99-966d-c763d22081e2@nvidia.com/ Tested-by: Jon Hunter Acked-by: Jon Hunter Link: https://patch.msgid.link/20260318-gpio-shared-xlate-v2-2-0ce34c707e81@oss.qualcomm.com Signed-off-by: Bartosz Golaszewski

gpio: shared: call gpio_chip::of_xlate() if set

2026-03-23T08:54:24+00:00

OF-based GPIO controller drivers may provide a translation function that calculates the real chip offset from whatever devicetree sources provide. We need to take this into account in the shared GPIO management and call of_xlate() if it's provided and adjust the entry->offset we initially set when scanning the tree. To that end: modify the shared GPIO API to take the GPIO chip as argument on setup (to avoid having to rcu_dereference() it from the GPIO device) and protect the access to entry->offset with the existing lock. Fixes: a060b8c511ab ("gpiolib: implement low-level, shared GPIO support") Reported-by: Jon Hunter Closes: https://lore.kernel.org/all/921ba8ce-b18e-4a99-966d-c763d22081e2@nvidia.com/ Reviewed-by: Linus Walleij Tested-by: Jon Hunter Acked-by: Jon Hunter Link: https://patch.msgid.link/20260318-gpio-shared-xlate-v2-1-0ce34c707e81@oss.qualcomm.com Signed-off-by: Bartosz Golaszewski

Merge branch 'gpio/dev-init-rework' into gpio/for-current

2026-03-16T09:23:47+00:00

Pull in the gpiochip_add_data_with_key() rework addressing resource leaks in error path.

Revert "gpio: Access `gpio_bus_type` in gpiochip_setup_dev()"

2026-03-16T09:10:02+00:00

This reverts commit cc11f4ef666fbca02c8a2f11d0184d57e6b75579. Commit cc11f4ef666f ("gpio: Access `gpio_bus_type` in gpiochip_setup_dev()") moved the bus type assignment from gpiochip_add_data_with_key() to gpiochip_setup_dev(). This change introduced a bug where dev_printk() and friends might access the bus name after gpiochip_add_data_with_key() but before gpiochip_setup_dev() has run. In this window, the bus type is not yet initialized, leading to empty bus names in logs. Move the bus type assignment back to gpiochip_add_data_with_key() to ensure the bus name is available before any potential users like dev_printk(). Reported-by: Geert Uytterhoeven Closes: https://lore.kernel.org/all/CAMuHMdU0Xb=Moca5LUex+VxyHQa2-uYJgYf4hzHiSEjDCQQT=Q@mail.gmail.com/ Signed-off-by: Tzung-Bi Shih Link: https://patch.msgid.link/20260313054112.1248074-1-tzungbi@kernel.org Signed-off-by: Bartosz Golaszewski

gpiolib: Update gpiochip_find_base_unlocked() kerneldoc

2026-03-16T08:53:25+00:00

This function albeit static was lacking a kerneldoc, and the function returns a dangerous number for internal use so make that clear in the kerneldoc. Reported-by: Matthijs Kooijman Signed-off-by: Linus Walleij Link: https://patch.msgid.link/20260311-gpio-discourage-dynamic-v1-1-c8b68fc84203@kernel.org Signed-off-by: Bartosz Golaszewski