linux-stable.git/drivers/pinctrl/qcom, branch v6.7 Linux kernel stable tree Merge tag 'pinctrl-v6.7-1' of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-pinctrl 2023-11-04T05:15:19+00:00 Linus Torvalds torvalds@linux-foundation.org 2023-11-04T05:15:19+00:00 90b0c2b2edd1adff742c621e246562fbefa11b70 Pull pin control updates from Linus Walleij: "No pin control core changes this time. New drivers: - Realtek RTD family pin control driver and RTD1619B, RTD1319D and RTD1315E subdrivers - Nuvoton NPCM8xx combined pin control and GPIO driver - Amlogic T7 pin control driver - Renesas RZ/G3S pin control driver Improvements: - A number of additional UART groups added to the Mediatek MT7981 driver - MPM pin maps added for Qualcomm MSM8996, SM6115, SM6125 and SDM660 - Extra GPIO banks for the Sunxi H616 - MLSP I2C6 function support in Qualcomm MSM8226 - Some __counted_by() annotations for dynamic arrays - Ongoing work to make remove() return void - LSBC groups and functions in the Renesas R8A7778" * tag 'pinctrl-v6.7-1' of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-pinctrl: (110 commits) pinctrl: Use device_get_match_data() dt-bindings: pinctrl: qcom,sa8775p-tlmm: add missing wakeup-parent dt-bindings: pinctrl: nuvoton,npcm845: Add missing additionalProperties on gpio child nodes dt-bindings: pinctrl: brcm: Ensure all child node properties are documented pinctrl: renesas: rzn1: Convert to platform remove callback returning void pinctrl: renesas: rzg2l: Add RZ/G3S support dt-bindings: pinctrl: renesas: Document RZ/G3S SoC pinctrl: renesas: rzg2l: Add support for different DS values on different groups pinctrl: renesas: rzg2l: Move DS and OI to SoC-specific configuration pinctrl: renesas: rzg2l: Adapt function number for RZ/G3S pinctrl: renesas: rzg2l: Adapt for different SD/PWPR register offsets pinctrl: renesas: rzg2l: Index all registers based on port offset pinctrl: renesas: rzg2l: Add validation of GPIO pin in rzg2l_gpio_request() pinctrl: renesas: r8a7778: Add LBSC pins, groups, and functions pinctrl: intel: fetch community only when we need it pinctrl: cherryview: reduce scope of PIN_CONFIG_BIAS_HIGH_IMPEDANCE case pinctrl: cherryview: Convert to platform remove callback returning void pinctrl: sprd-sc9860: Convert to platform remove callback returning void pinctrl: qcom/msm: Convert to platform remove callback returning void pinctrl: qcom/lpi: Convert to platform remove callback returning void ... 
Pull pin control updates from Linus Walleij:
 "No pin control core changes this time.

  New drivers:

   - Realtek RTD family pin control driver and RTD1619B, RTD1319D and
     RTD1315E subdrivers

   - Nuvoton NPCM8xx combined pin control and GPIO driver

   - Amlogic T7 pin control driver

   - Renesas RZ/G3S pin control driver

  Improvements:

   - A number of additional UART groups added to the Mediatek MT7981
     driver

   - MPM pin maps added for Qualcomm MSM8996, SM6115, SM6125 and SDM660

   - Extra GPIO banks for the Sunxi H616

   - MLSP I2C6 function support in Qualcomm MSM8226

   - Some __counted_by() annotations for dynamic arrays

   - Ongoing work to make remove() return void

   - LSBC groups and functions in the Renesas R8A7778"

* tag 'pinctrl-v6.7-1' of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-pinctrl: (110 commits)
  pinctrl: Use device_get_match_data()
  dt-bindings: pinctrl: qcom,sa8775p-tlmm: add missing wakeup-parent
  dt-bindings: pinctrl: nuvoton,npcm845: Add missing additionalProperties on gpio child nodes
  dt-bindings: pinctrl: brcm: Ensure all child node properties are documented
  pinctrl: renesas: rzn1: Convert to platform remove callback returning void
  pinctrl: renesas: rzg2l: Add RZ/G3S support
  dt-bindings: pinctrl: renesas: Document RZ/G3S SoC
  pinctrl: renesas: rzg2l: Add support for different DS values on different groups
  pinctrl: renesas: rzg2l: Move DS and OI to SoC-specific configuration
  pinctrl: renesas: rzg2l: Adapt function number for RZ/G3S
  pinctrl: renesas: rzg2l: Adapt for different SD/PWPR register offsets
  pinctrl: renesas: rzg2l: Index all registers based on port offset
  pinctrl: renesas: rzg2l: Add validation of GPIO pin in rzg2l_gpio_request()
  pinctrl: renesas: r8a7778: Add LBSC pins, groups, and functions
  pinctrl: intel: fetch community only when we need it
  pinctrl: cherryview: reduce scope of PIN_CONFIG_BIAS_HIGH_IMPEDANCE case
  pinctrl: cherryview: Convert to platform remove callback returning void
  pinctrl: sprd-sc9860: Convert to platform remove callback returning void
  pinctrl: qcom/msm: Convert to platform remove callback returning void
  pinctrl: qcom/lpi: Convert to platform remove callback returning void
  ...
pinctrl: qcom: lpass-lpi: fix concurrent register updates 2023-10-16T12:56:49+00:00 Krzysztof Kozlowski krzysztof.kozlowski@linaro.org 2023-10-13T14:57:05+00:00 c8befdc411e5fd1bf95a13e8744c8ca79b412bee The Qualcomm LPASS LPI pin controller driver uses one lock for guarding Read-Modify-Write code for slew rate registers. However the pin configuration and muxing registers have exactly the same RMW code but are not protected. Pin controller framework does not provide locking here, thus it is possible to trigger simultaneous change of pin configuration registers resulting in non-atomic changes. Protect from concurrent access by re-using the same lock used to cover the slew rate register. Using the same lock instead of adding second one will make more sense, once we add support for newer Qualcomm SoC, where slew rate is configured in the same register as pin configuration/muxing. Fixes: 6e261d1090d6 ("pinctrl: qcom: Add sm8250 lpass lpi pinctrl driver") Cc: stable@vger.kernel.org Reviewed-by: Linus Walleij <linus.walleij@linaro.org> Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org> Link: https://lore.kernel.org/r/20231013145705.219954-1-krzysztof.kozlowski@linaro.org Signed-off-by: Linus Walleij <linus.walleij@linaro.org> 
The Qualcomm LPASS LPI pin controller driver uses one lock for guarding
Read-Modify-Write code for slew rate registers.  However the pin
configuration and muxing registers have exactly the same RMW code but
are not protected.

Pin controller framework does not provide locking here, thus it is
possible to trigger simultaneous change of pin configuration registers
resulting in non-atomic changes.

Protect from concurrent access by re-using the same lock used to cover
the slew rate register.  Using the same lock instead of adding second
one will make more sense, once we add support for newer Qualcomm SoC,
where slew rate is configured in the same register as pin
configuration/muxing.

Fixes: 6e261d1090d6 ("pinctrl: qcom: Add sm8250 lpass lpi pinctrl driver")
Cc: stable@vger.kernel.org
Reviewed-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Link: https://lore.kernel.org/r/20231013145705.219954-1-krzysztof.kozlowski@linaro.org
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
pinctrl: qcom/msm: Convert to platform remove callback returning void 2023-10-10T13:47:49+00:00 Uwe Kleine-König u.kleine-koenig@pengutronix.de 2023-10-09T16:25:09+00:00 22ee670a8ad3ec7cd9d872d4512fe8797130e191 The .remove() callback for a platform driver returns an int which makes many driver authors wrongly assume it's possible to do error handling by returning an error code. However the value returned is ignored (apart from emitting a warning) and this typically results in resource leaks. To improve here there is a quest to make the remove callback return void. In the first step of this quest all drivers are converted to .remove_new(), which already returns void. Eventually after all drivers are converted, .remove_new() will be renamed to .remove(). To convert all those qcom pinctrl drivers, make msm_pinctrl_remove() return void (instead of zero) and use .remove_new in all drivers. Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de> Link: https://lore.kernel.org/r/20231009162510.335208-3-u.kleine-koenig@pengutronix.de Signed-off-by: Linus Walleij <linus.walleij@linaro.org> 
The .remove() callback for a platform driver returns an int which makes
many driver authors wrongly assume it's possible to do error handling by
returning an error code. However the value returned is ignored (apart
from emitting a warning) and this typically results in resource leaks.

To improve here there is a quest to make the remove callback return
void. In the first step of this quest all drivers are converted to
.remove_new(), which already returns void. Eventually after all drivers
are converted, .remove_new() will be renamed to .remove().

To convert all those qcom pinctrl drivers, make msm_pinctrl_remove()
return void (instead of zero) and use .remove_new in all drivers.

Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Link: https://lore.kernel.org/r/20231009162510.335208-3-u.kleine-koenig@pengutronix.de
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
pinctrl: qcom/lpi: Convert to platform remove callback returning void 2023-10-10T13:47:49+00:00 Uwe Kleine-König u.kleine-koenig@pengutronix.de 2023-10-09T16:25:08+00:00 776d54c137a38f99701fa5cc65d947244dda1b78 The .remove() callback for a platform driver returns an int which makes many driver authors wrongly assume it's possible to do error handling by returning an error code. However the value returned is ignored (apart from emitting a warning) and this typically results in resource leaks. To improve here there is a quest to make the remove callback return void. In the first step of this quest all drivers are converted to .remove_new(), which already returns void. Eventually after all drivers are converted, .remove_new() will be renamed to .remove(). To convert all those qcom pinctrl drivers, make msm_pinctrl_remove() return void (instead of zero) and use .remove_new in all drivers. Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de> Link: https://lore.kernel.org/r/20231009162510.335208-2-u.kleine-koenig@pengutronix.de Signed-off-by: Linus Walleij <linus.walleij@linaro.org> 
The .remove() callback for a platform driver returns an int which makes
many driver authors wrongly assume it's possible to do error handling by
returning an error code. However the value returned is ignored (apart
from emitting a warning) and this typically results in resource leaks.

To improve here there is a quest to make the remove callback return
void. In the first step of this quest all drivers are converted to
.remove_new(), which already returns void. Eventually after all drivers
are converted, .remove_new() will be renamed to .remove().

To convert all those qcom pinctrl drivers, make msm_pinctrl_remove()
return void (instead of zero) and use .remove_new in all drivers.

Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Link: https://lore.kernel.org/r/20231009162510.335208-2-u.kleine-koenig@pengutronix.de
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
pinctrl: qcom: ssbi-mpp: Convert to platform remove callback returning void 2023-10-10T13:43:56+00:00 Uwe Kleine-König u.kleine-koenig@pengutronix.de 2023-10-09T08:38:54+00:00 a80252efb5d539b292433189f282f80aabaa53b1 The .remove() callback for a platform driver returns an int which makes many driver authors wrongly assume it's possible to do error handling by returning an error code. However the value returned is ignored (apart from emitting a warning) and this typically results in resource leaks. To improve here there is a quest to make the remove callback return void. In the first step of this quest all drivers are converted to .remove_new(), which already returns void. Eventually after all drivers are converted, .remove_new() will be renamed to .remove(). Trivially convert this driver from always returning zero in the remove callback to the void returning variant. Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de> Link: https://lore.kernel.org/r/20231009083856.222030-19-u.kleine-koenig@pengutronix.de Signed-off-by: Linus Walleij <linus.walleij@linaro.org> 
The .remove() callback for a platform driver returns an int which makes
many driver authors wrongly assume it's possible to do error handling by
returning an error code. However the value returned is ignored (apart
from emitting a warning) and this typically results in resource leaks.

To improve here there is a quest to make the remove callback return
void. In the first step of this quest all drivers are converted to
.remove_new(), which already returns void. Eventually after all drivers
are converted, .remove_new() will be renamed to .remove().

Trivially convert this driver from always returning zero in the remove
callback to the void returning variant.

Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Link: https://lore.kernel.org/r/20231009083856.222030-19-u.kleine-koenig@pengutronix.de
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
pinctrl: qcom: ssbi-gpio: Convert to platform remove callback returning void 2023-10-10T13:43:31+00:00 Uwe Kleine-König u.kleine-koenig@pengutronix.de 2023-10-09T08:38:53+00:00 a37b06133aa426f83d8f6497785dd93b30a5bc5e The .remove() callback for a platform driver returns an int which makes many driver authors wrongly assume it's possible to do error handling by returning an error code. However the value returned is ignored (apart from emitting a warning) and this typically results in resource leaks. To improve here there is a quest to make the remove callback return void. In the first step of this quest all drivers are converted to .remove_new(), which already returns void. Eventually after all drivers are converted, .remove_new() will be renamed to .remove(). Trivially convert this driver from always returning zero in the remove callback to the void returning variant. Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de> Link: https://lore.kernel.org/r/20231009083856.222030-18-u.kleine-koenig@pengutronix.de Signed-off-by: Linus Walleij <linus.walleij@linaro.org> 
The .remove() callback for a platform driver returns an int which makes
many driver authors wrongly assume it's possible to do error handling by
returning an error code. However the value returned is ignored (apart
from emitting a warning) and this typically results in resource leaks.

To improve here there is a quest to make the remove callback return
void. In the first step of this quest all drivers are converted to
.remove_new(), which already returns void. Eventually after all drivers
are converted, .remove_new() will be renamed to .remove().

Trivially convert this driver from always returning zero in the remove
callback to the void returning variant.

Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Link: https://lore.kernel.org/r/20231009083856.222030-18-u.kleine-koenig@pengutronix.de
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
pinctrl: qcom: spmi-mpp: Convert to platform remove callback returning void 2023-10-10T13:43:03+00:00 Uwe Kleine-König u.kleine-koenig@pengutronix.de 2023-10-09T08:38:52+00:00 ab06ab84065facb985b340871b3bde957298c156 The .remove() callback for a platform driver returns an int which makes many driver authors wrongly assume it's possible to do error handling by returning an error code. However the value returned is ignored (apart from emitting a warning) and this typically results in resource leaks. To improve here there is a quest to make the remove callback return void. In the first step of this quest all drivers are converted to .remove_new(), which already returns void. Eventually after all drivers are converted, .remove_new() will be renamed to .remove(). Trivially convert this driver from always returning zero in the remove callback to the void returning variant. Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de> Link: https://lore.kernel.org/r/20231009083856.222030-17-u.kleine-koenig@pengutronix.de Signed-off-by: Linus Walleij <linus.walleij@linaro.org> 
The .remove() callback for a platform driver returns an int which makes
many driver authors wrongly assume it's possible to do error handling by
returning an error code. However the value returned is ignored (apart
from emitting a warning) and this typically results in resource leaks.

To improve here there is a quest to make the remove callback return
void. In the first step of this quest all drivers are converted to
.remove_new(), which already returns void. Eventually after all drivers
are converted, .remove_new() will be renamed to .remove().

Trivially convert this driver from always returning zero in the remove
callback to the void returning variant.

Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Link: https://lore.kernel.org/r/20231009083856.222030-17-u.kleine-koenig@pengutronix.de
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
pinctrl: qcom: spmi-gpio: Convert to platform remove callback returning void 2023-10-10T13:42:37+00:00 Uwe Kleine-König u.kleine-koenig@pengutronix.de 2023-10-09T08:38:51+00:00 348e1488ba5cd55f5f7d96e5ab3672b44dc1adb0 The .remove() callback for a platform driver returns an int which makes many driver authors wrongly assume it's possible to do error handling by returning an error code. However the value returned is ignored (apart from emitting a warning) and this typically results in resource leaks. To improve here there is a quest to make the remove callback return void. In the first step of this quest all drivers are converted to .remove_new(), which already returns void. Eventually after all drivers are converted, .remove_new() will be renamed to .remove(). Trivially convert this driver from always returning zero in the remove callback to the void returning variant. Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de> Link: https://lore.kernel.org/r/20231009083856.222030-16-u.kleine-koenig@pengutronix.de Signed-off-by: Linus Walleij <linus.walleij@linaro.org> 
The .remove() callback for a platform driver returns an int which makes
many driver authors wrongly assume it's possible to do error handling by
returning an error code. However the value returned is ignored (apart
from emitting a warning) and this typically results in resource leaks.

To improve here there is a quest to make the remove callback return
void. In the first step of this quest all drivers are converted to
.remove_new(), which already returns void. Eventually after all drivers
are converted, .remove_new() will be renamed to .remove().

Trivially convert this driver from always returning zero in the remove
callback to the void returning variant.

Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Link: https://lore.kernel.org/r/20231009083856.222030-16-u.kleine-koenig@pengutronix.de
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
pinctrl: qcom: msm8974: Add MPM pin mappings 2023-09-27T08:43:51+00:00 Matti Lehtimäki matti.lehtimaki@gmail.com 2023-09-23T13:14:32+00:00 1d4b7855b1cf7a00a46bb8158572727c63f50a22 Add pin <-> wakeirq mappings to allow for waking up the AP from sleep through MPM-connected pins. Signed-off-by: Matti Lehtimäki <matti.lehtimaki@gmail.com> Reviewed-by: Luca Weiss <luca@z3ntu.xyz> Link: https://lore.kernel.org/r/20230923131432.21721-3-matti.lehtimaki@gmail.com Signed-off-by: Linus Walleij <linus.walleij@linaro.org> 
Add pin <-> wakeirq mappings to allow for waking up the AP from sleep
through MPM-connected pins.

Signed-off-by: Matti Lehtimäki <matti.lehtimaki@gmail.com>
Reviewed-by: Luca Weiss <luca@z3ntu.xyz>
Link: https://lore.kernel.org/r/20230923131432.21721-3-matti.lehtimaki@gmail.com
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
pinctrl: qcom: msm8226: Add MPM pin mappings 2023-09-27T08:43:51+00:00 Matti Lehtimäki matti.lehtimaki@gmail.com 2023-09-23T13:14:31+00:00 259fe3cbf2a537e6534bdcff314555f0eb77a886 Add pin <-> wakeirq mappings to allow for waking up the AP from sleep through MPM-connected pins. Signed-off-by: Matti Lehtimäki <matti.lehtimaki@gmail.com> Reviewed-by: Luca Weiss <luca@z3ntu.xyz> Link: https://lore.kernel.org/r/20230923131432.21721-2-matti.lehtimaki@gmail.com Signed-off-by: Linus Walleij <linus.walleij@linaro.org> 
Add pin <-> wakeirq mappings to allow for waking up the AP from sleep
through MPM-connected pins.

Signed-off-by: Matti Lehtimäki <matti.lehtimaki@gmail.com>
Reviewed-by: Luca Weiss <luca@z3ntu.xyz>
Link: https://lore.kernel.org/r/20230923131432.21721-2-matti.lehtimaki@gmail.com
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>