linux.git/drivers/gpio, branch v7.1-rc5 Linux kernel source tree gpio: virtuser: lock device when calling device_is_bound() 2026-05-21T12:36:07+00:00 Bartosz Golaszewski bartosz.golaszewski@oss.qualcomm.com 2026-05-18T09:53:19+00:00 a4fa45c1d980bc2b9837f469119af24a9304a1fc The kerneldoc for device_is_bound() says it must be called with the device lock taken. Add missing synchronization to this driver. Fixes: c3e2a8aef28c ("gpio: virtuser: stop using dev-sync-probe") Link: https://patch.msgid.link/20260518-gpio-dev-lock-v1-3-cc4736f3ff0b@oss.qualcomm.com Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@oss.qualcomm.com> 
The kerneldoc for device_is_bound() says it must be called with the
device lock taken. Add missing synchronization to this driver.

Fixes: c3e2a8aef28c ("gpio: virtuser: stop using dev-sync-probe")
Link: https://patch.msgid.link/20260518-gpio-dev-lock-v1-3-cc4736f3ff0b@oss.qualcomm.com
Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@oss.qualcomm.com>
gpio: aggregator: lock device when calling device_is_bound() 2026-05-21T12:36:06+00:00 Bartosz Golaszewski bartosz.golaszewski@oss.qualcomm.com 2026-05-18T09:53:18+00:00 598a2b3e2e0e6aa2e9f7843c96c45b5ea11e0411 The kerneldoc for device_is_bound() says it must be called with the device lock taken. Add missing synchronization to this driver. Fixes: 3a27f40b4570 ("gpio: aggregator: stop using dev-sync-probe") Link: https://patch.msgid.link/20260518-gpio-dev-lock-v1-2-cc4736f3ff0b@oss.qualcomm.com Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@oss.qualcomm.com> 
The kerneldoc for device_is_bound() says it must be called with the
device lock taken. Add missing synchronization to this driver.

Fixes: 3a27f40b4570 ("gpio: aggregator: stop using dev-sync-probe")
Link: https://patch.msgid.link/20260518-gpio-dev-lock-v1-2-cc4736f3ff0b@oss.qualcomm.com
Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@oss.qualcomm.com>
gpio: sim: lock device when calling device_is_bound() 2026-05-21T12:36:06+00:00 Bartosz Golaszewski bartosz.golaszewski@oss.qualcomm.com 2026-05-18T09:53:17+00:00 e70ae40d6660c5428c790c329318c570b4d038ab The kerneldoc for device_is_bound() says it must be called with the device lock taken. Add missing synchronization to this driver. Fixes: 7fb3287946f9 ("gpio: sim: stop using dev-sync-probe") Link: https://patch.msgid.link/20260518-gpio-dev-lock-v1-1-cc4736f3ff0b@oss.qualcomm.com Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@oss.qualcomm.com> 
The kerneldoc for device_is_bound() says it must be called with the
device lock taken. Add missing synchronization to this driver.

Fixes: 7fb3287946f9 ("gpio: sim: stop using dev-sync-probe")
Link: https://patch.msgid.link/20260518-gpio-dev-lock-v1-1-cc4736f3ff0b@oss.qualcomm.com
Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@oss.qualcomm.com>
gpio: aggregator: remove the software node when deactivating the aggregator 2026-05-21T10:01:07+00:00 Bartosz Golaszewski bartosz.golaszewski@oss.qualcomm.com 2026-05-20T12:16:31+00:00 61fef83f239ecace1cce716135762a2d9b7b1fc6 The dynamic software node we create for the aggregator platform device when using configfs is leaked when the device is deactivated. Destroy it as the last step in the tear-down path. Fixes: 86f162e73d2d ("gpio: aggregator: introduce basic configfs interface") Reported-by: Geert Uytterhoeven <geert@linux-m68k.org> Closes: https://lore.kernel.org/all/CAMuHMdVZ=XUvJTGdDAjnkxgtw7Uvnn61iOy3XN_5XNZM2anctw@mail.gmail.com/ Link: https://patch.msgid.link/20260520121631.33976-1-bartosz.golaszewski@oss.qualcomm.com Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@oss.qualcomm.com> 
The dynamic software node we create for the aggregator platform device
when using configfs is leaked when the device is deactivated. Destroy it
as the last step in the tear-down path.

Fixes: 86f162e73d2d ("gpio: aggregator: introduce basic configfs interface")
Reported-by: Geert Uytterhoeven <geert@linux-m68k.org>
Closes: https://lore.kernel.org/all/CAMuHMdVZ=XUvJTGdDAjnkxgtw7Uvnn61iOy3XN_5XNZM2anctw@mail.gmail.com/
Link: https://patch.msgid.link/20260520121631.33976-1-bartosz.golaszewski@oss.qualcomm.com
Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@oss.qualcomm.com>
gpio: aggregator: fix a potential use-after-free 2026-05-21T10:00:42+00:00 Bartosz Golaszewski bartosz.golaszewski@oss.qualcomm.com 2026-05-20T08:49:11+00:00 30c073cab97afb31901f94de9605177b6b84367e On error we free aggr->lookups->dev_id before removing the entry from the lookup table. If a concurrent thread calls gpiod_find() before we remove the entry, it could iterate over the list and call gpiod_match_lookup_table() which unconditionally dereferences dev_id when calling strcmp(). Reverse the order of cleanup. Fixes: 86f162e73d2d ("gpio: aggregator: introduce basic configfs interface") Reviewed-by: Geert Uytterhoeven <geert+renesas@glider.be> Link: https://patch.msgid.link/20260520084911.27938-1-bartosz.golaszewski@oss.qualcomm.com Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@oss.qualcomm.com> 
On error we free aggr->lookups->dev_id before removing the entry from
the lookup table. If a concurrent thread calls gpiod_find() before we
remove the entry, it could iterate over the list and call
gpiod_match_lookup_table() which unconditionally dereferences dev_id
when calling strcmp(). Reverse the order of cleanup.

Fixes: 86f162e73d2d ("gpio: aggregator: introduce basic configfs interface")
Reviewed-by: Geert Uytterhoeven <geert+renesas@glider.be>
Link: https://patch.msgid.link/20260520084911.27938-1-bartosz.golaszewski@oss.qualcomm.com
Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@oss.qualcomm.com>
gpio: cdev: check if uAPI v2 config attributes are correctly zeroed 2026-05-21T10:00:04+00:00 Bartosz Golaszewski bartosz.golaszewski@oss.qualcomm.com 2026-05-21T08:42:16+00:00 3e6ccd790ed69bedd3d9626d01dd35cf9821c121 We check the padding of other uAPI v2 structures but not that of line config attributes. For used attributes: check if their padding is zeroed, for unused: check if the entire structure is zeroed. Fixes: 3c0d9c635ae2 ("gpiolib: cdev: support GPIO_V2_GET_LINE_IOCTL and GPIO_V2_LINE_GET_VALUES_IOCTL") Reviewed-by: Kent Gibson <warthog618@gmail.com> Link: https://patch.msgid.link/20260521-gpio-cdev-attr-padding-check-v3-1-ec3bcbe2e358@oss.qualcomm.com Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@oss.qualcomm.com> 
We check the padding of other uAPI v2 structures but not that of line
config attributes. For used attributes: check if their padding is
zeroed, for unused: check if the entire structure is zeroed.

Fixes: 3c0d9c635ae2 ("gpiolib: cdev: support GPIO_V2_GET_LINE_IOCTL and GPIO_V2_LINE_GET_VALUES_IOCTL")
Reviewed-by: Kent Gibson <warthog618@gmail.com>
Link: https://patch.msgid.link/20260521-gpio-cdev-attr-padding-check-v3-1-ec3bcbe2e358@oss.qualcomm.com
Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@oss.qualcomm.com>
gpio: pca953x: propagate regulator_enable() error from resume 2026-05-21T08:04:00+00:00 Stepan Ionichev sozdayvek@gmail.com 2026-05-20T11:05:04+00:00 be460cedb67ab803c1bebceac19b1d44acb85d30 pca953x_resume() returns 0 when regulator_enable() fails, dropping the real error code and masking the failure as a successful resume. The caller then proceeds as if the chip is powered, while the regulator is in fact disabled. Return ret so PM core sees the actual failure. Signed-off-by: Stepan Ionichev <sozdayvek@gmail.com> Link: https://patch.msgid.link/20260520110504.13969-1-sozdayvek@gmail.com Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@oss.qualcomm.com> 
pca953x_resume() returns 0 when regulator_enable() fails, dropping
the real error code and masking the failure as a successful resume.
The caller then proceeds as if the chip is powered, while the
regulator is in fact disabled.

Return ret so PM core sees the actual failure.

Signed-off-by: Stepan Ionichev <sozdayvek@gmail.com>
Link: https://patch.msgid.link/20260520110504.13969-1-sozdayvek@gmail.com
Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@oss.qualcomm.com>
Merge tag 'gpio-fixes-for-v7.1-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/brgl/linux 2026-04-24T18:59:46+00:00 Linus Torvalds torvalds@linux-foundation.org 2026-04-24T18:59:46+00:00 d0fc5bf9fe9b89389287846f13cc4e462a89954d Pull gpio fixes from Bartosz Golaszewski: - fix a regression in gpio-rockchip introduced on older chips during the merge window when converting to dynamic GPIO base - fix AST2700 debounce selector bit definitions in gpio-aspeed * tag 'gpio-fixes-for-v7.1-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/brgl/linux: gpio: aspeed: fix AST2700 debounce selector bit definitions gpio: rockchip: Fix GPIO regression after conversion to dynamic base allocation 
Pull gpio fixes from Bartosz Golaszewski:

 - fix a regression in gpio-rockchip introduced on older chips during
   the merge window when converting to dynamic GPIO base

 - fix AST2700 debounce selector bit definitions in gpio-aspeed

* tag 'gpio-fixes-for-v7.1-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/brgl/linux:
  gpio: aspeed: fix AST2700 debounce selector bit definitions
  gpio: rockchip: Fix GPIO regression after conversion to dynamic base allocation
gpio: aspeed: fix AST2700 debounce selector bit definitions 2026-04-20T09:10:39+00:00 Billy Tsai billy_tsai@aspeedtech.com 2026-04-15T10:24:42+00:00 e31eee4a961077d60ef2362507240c6743c1c2ae The AST2700 datasheet defines reg_debounce_sel1 as the low bit and reg_debounce_sel2 as the high bit. The current driver uses the AST2600 mapping instead, where sel1 is the high bit and sel2 is the low bit. As a result, the debounce selector bits are programmed in reverse on AST2700. Swap the G7 sel1/sel2 bit definitions so the driver matches the hardware definition. Fixes: b2e861bd1eaf ("gpio: aspeed: Support G7 Aspeed gpio controller") Signed-off-by: Billy Tsai <billy_tsai@aspeedtech.com> Link: https://patch.msgid.link/20260415-gpio-fix-v1-1-b08a89b31e6f@aspeedtech.com Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@oss.qualcomm.com> 
The AST2700 datasheet defines reg_debounce_sel1 as the low bit and
reg_debounce_sel2 as the high bit. The current driver uses the AST2600
mapping instead, where sel1 is the high bit and sel2 is the low bit.

As a result, the debounce selector bits are programmed in reverse on
AST2700. Swap the G7 sel1/sel2 bit definitions so the driver matches the
hardware definition.

Fixes: b2e861bd1eaf ("gpio: aspeed: Support G7 Aspeed gpio controller")
Signed-off-by: Billy Tsai <billy_tsai@aspeedtech.com>
Link: https://patch.msgid.link/20260415-gpio-fix-v1-1-b08a89b31e6f@aspeedtech.com
Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@oss.qualcomm.com>
gpio: rockchip: Fix GPIO regression after conversion to dynamic base allocation 2026-04-20T08:46:31+00:00 Jonas Karlman jonas@kwiboo.se 2026-04-16T15:49:28+00:00 5cd9c6d332f46d1de8b68117fe2a3f1b08ee80ff The commit c8079f83e0bf ("gpio: rockchip: convert to dynamic GPIO base allocation") broke GPIO on devices using device trees which don't set the gpio-ranges property, something only Rockchip RK35xx SoC DTs do. On a Rockchip RK3399 device something like following is now observed: [ 0.082771] rockchip-gpio ff720000.gpio: probed /pinctrl/gpio@ff720000 [ 0.083531] rockchip-gpio ff730000.gpio: probed /pinctrl/gpio@ff730000 [ 0.084110] rockchip-gpio ff780000.gpio: probed /pinctrl/gpio@ff780000 [ 0.084746] rockchip-gpio ff788000.gpio: probed /pinctrl/gpio@ff788000 [ 0.085389] rockchip-gpio ff790000.gpio: probed /pinctrl/gpio@ff790000 -- [ 0.212208] rockchip-pinctrl pinctrl: pin 637 is not registered so it cannot be requested [ 0.212271] rockchip-pinctrl pinctrl: error -EINVAL: pin-637 (gpio3:637) [ 0.212344] leds-gpio leds: error -EINVAL: Failed to get GPIO '/leds/led-0' [ 0.212389] leds-gpio leds: probe with driver leds-gpio failed with error -22 -- [ 0.607545] rockchip-pinctrl pinctrl: pin 519 is not registered so it cannot be requested [ 0.608775] rockchip-pinctrl pinctrl: error -EINVAL: pin-519 (gpio0:519) [ 0.610003] dwmmc_rockchip fe320000.mmc: probe with driver dwmmc_rockchip failed with error -22 -- [ 0.805882] rockchip-pinctrl pinctrl: pin 547 is not registered so it cannot be requested [ 0.806672] rockchip-pinctrl pinctrl: error -EINVAL: pin-547 (gpio1:547) [ 0.807301] reg-fixed-voltage regulator-vbus-typec: error -EINVAL: can't get GPIO [ 0.807307] rockchip-pinctrl pinctrl: pin 602 is not registered so it cannot be requested [ 0.807970] reg-fixed-voltage regulator-vbus-typec: probe with driver reg-fixed-voltage failed with error -22 [ 0.808692] rockchip-pinctrl pinctrl: error -EINVAL: pin-602 (gpio2:602) [ 0.810279] reg-fixed-voltage regulator-vcc3v3-pcie: error -EINVAL: can't get GPIO [ 0.810284] rockchip-pinctrl pinctrl: pin 665 is not registered so it cannot be requested [ 0.810299] rockchip-pinctrl pinctrl: error -EINVAL: pin-665 (gpio4:665) [ 0.810960] reg-fixed-voltage regulator-vcc3v3-pcie: probe with driver reg-fixed-voltage failed with error -22 [ 0.811679] reg-fixed-voltage regulator-vcc5v0-host: error -EINVAL: can't get GPIO [ 0.813943] reg-fixed-voltage regulator-vcc5v0-host: probe with driver reg-fixed-voltage failed with error -22 -- [ 0.867788] rockchip-pinctrl pinctrl: pin 522 is not registered so it cannot be requested [ 0.868537] rockchip-pinctrl pinctrl: error -EINVAL: pin-522 (gpio0:522) [ 0.869166] pwrseq_simple sdio-pwrseq: error -EINVAL: reset GPIOs not ready [ 0.869798] pwrseq_simple sdio-pwrseq: probe with driver pwrseq_simple failed with error -22 -- [ 0.940365] rockchip-pinctrl pinctrl: pin 623 is not registered so it cannot be requested [ 0.941084] rockchip-pinctrl pinctrl: error -EINVAL: pin-623 (gpio3:623) [ 0.941823] rk_gmac-dwmac fe300000.ethernet: error -EINVAL: Cannot register the MDIO bus [ 0.942542] rk_gmac-dwmac fe300000.ethernet: error -EINVAL: MDIO bus (id: 0) registration failed [ 0.943772] rk_gmac-dwmac fe300000.ethernet: probe with driver rk_gmac-dwmac failed with error -22 Restore GPIO to a working state on devices using older Rockchip SoCs and/or DTs not having the gpio-ranges property set by restoring prior use of bank->pin_base as the pin_offset value. Also change to use bank->nr_pins as the npins value to align and prevent a possible future breakage if gc->ngpio is ever changed to match the 32 GPIOs each controller theoretically can handle. Fixes: c8079f83e0bf ("gpio: rockchip: convert to dynamic GPIO base allocation") Signed-off-by: Jonas Karlman <jonas@kwiboo.se> Reviewed-by: Linus Walleij <linusw@kernel.org> Acked-by: Heiko Stuebner <heiko@sntech.de> Link: https://patch.msgid.link/20260416154928.2103388-1-jonas@kwiboo.se Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@oss.qualcomm.com> 
The commit c8079f83e0bf ("gpio: rockchip: convert to dynamic GPIO base
allocation") broke GPIO on devices using device trees which don't set
the gpio-ranges property, something only Rockchip RK35xx SoC DTs do.

On a Rockchip RK3399 device something like following is now observed:

[    0.082771] rockchip-gpio ff720000.gpio: probed /pinctrl/gpio@ff720000
[    0.083531] rockchip-gpio ff730000.gpio: probed /pinctrl/gpio@ff730000
[    0.084110] rockchip-gpio ff780000.gpio: probed /pinctrl/gpio@ff780000
[    0.084746] rockchip-gpio ff788000.gpio: probed /pinctrl/gpio@ff788000
[    0.085389] rockchip-gpio ff790000.gpio: probed /pinctrl/gpio@ff790000
--
[    0.212208] rockchip-pinctrl pinctrl: pin 637 is not registered so it cannot be requested
[    0.212271] rockchip-pinctrl pinctrl: error -EINVAL: pin-637 (gpio3:637)
[    0.212344] leds-gpio leds: error -EINVAL: Failed to get GPIO '/leds/led-0'
[    0.212389] leds-gpio leds: probe with driver leds-gpio failed with error -22
--
[    0.607545] rockchip-pinctrl pinctrl: pin 519 is not registered so it cannot be requested
[    0.608775] rockchip-pinctrl pinctrl: error -EINVAL: pin-519 (gpio0:519)
[    0.610003] dwmmc_rockchip fe320000.mmc: probe with driver dwmmc_rockchip failed with error -22
--
[    0.805882] rockchip-pinctrl pinctrl: pin 547 is not registered so it cannot be requested
[    0.806672] rockchip-pinctrl pinctrl: error -EINVAL: pin-547 (gpio1:547)
[    0.807301] reg-fixed-voltage regulator-vbus-typec: error -EINVAL: can't get GPIO
[    0.807307] rockchip-pinctrl pinctrl: pin 602 is not registered so it cannot be requested
[    0.807970] reg-fixed-voltage regulator-vbus-typec: probe with driver reg-fixed-voltage failed with error -22
[    0.808692] rockchip-pinctrl pinctrl: error -EINVAL: pin-602 (gpio2:602)
[    0.810279] reg-fixed-voltage regulator-vcc3v3-pcie: error -EINVAL: can't get GPIO
[    0.810284] rockchip-pinctrl pinctrl: pin 665 is not registered so it cannot be requested
[    0.810299] rockchip-pinctrl pinctrl: error -EINVAL: pin-665 (gpio4:665)
[    0.810960] reg-fixed-voltage regulator-vcc3v3-pcie: probe with driver reg-fixed-voltage failed with error -22
[    0.811679] reg-fixed-voltage regulator-vcc5v0-host: error -EINVAL: can't get GPIO
[    0.813943] reg-fixed-voltage regulator-vcc5v0-host: probe with driver reg-fixed-voltage failed with error -22
--
[    0.867788] rockchip-pinctrl pinctrl: pin 522 is not registered so it cannot be requested
[    0.868537] rockchip-pinctrl pinctrl: error -EINVAL: pin-522 (gpio0:522)
[    0.869166] pwrseq_simple sdio-pwrseq: error -EINVAL: reset GPIOs not ready
[    0.869798] pwrseq_simple sdio-pwrseq: probe with driver pwrseq_simple failed with error -22
--
[    0.940365] rockchip-pinctrl pinctrl: pin 623 is not registered so it cannot be requested
[    0.941084] rockchip-pinctrl pinctrl: error -EINVAL: pin-623 (gpio3:623)
[    0.941823] rk_gmac-dwmac fe300000.ethernet: error -EINVAL: Cannot register the MDIO bus
[    0.942542] rk_gmac-dwmac fe300000.ethernet: error -EINVAL: MDIO bus (id: 0) registration failed
[    0.943772] rk_gmac-dwmac fe300000.ethernet: probe with driver rk_gmac-dwmac failed with error -22

Restore GPIO to a working state on devices using older Rockchip SoCs
and/or DTs not having the gpio-ranges property set by restoring prior
use of bank->pin_base as the pin_offset value.

Also change to use bank->nr_pins as the npins value to align and prevent
a possible future breakage if gc->ngpio is ever changed to match the 32
GPIOs each controller theoretically can handle.

Fixes: c8079f83e0bf ("gpio: rockchip: convert to dynamic GPIO base allocation")
Signed-off-by: Jonas Karlman <jonas@kwiboo.se>
Reviewed-by: Linus Walleij <linusw@kernel.org>
Acked-by: Heiko Stuebner <heiko@sntech.de>
Link: https://patch.msgid.link/20260416154928.2103388-1-jonas@kwiboo.se
Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@oss.qualcomm.com>