wireguard-linux/drivers/pinctrl/pinctrl-single.c, branch update-toolchain

pinctrl: single: fix possible memory leak when pinctrl_enable() fails

2024-06-17T08:33:30Z

This driver calls pinctrl_register_and_init() which is not devm_ managed, it will leads memory leak if pinctrl_enable() fails. Replace it with devm_pinctrl_register_and_init(). And call pcs_free_resources() if pinctrl_enable() fails. Fixes: 5038a66dad01 ("pinctrl: core: delete incorrect free in pinctrl_enable()") Signed-off-by: Yang Yingliang Reviewed-by: Dan Carpenter Link: https://lore.kernel.org/r/20240606023704.3931561-3-yangyingliang@huawei.com Signed-off-by: Linus Walleij

pinctrl: single: Fix PIN_CONFIG_BIAS_DISABLE handling

2024-04-04T19:03:25Z

The pinctrl-single driver handles pin_config_set by looking up the requested setting in a DT-defined lookup table, which defines what bits correspond to each setting. There is no way to add PIN_CONFIG_BIAS_DISABLE entries to the table, since there is instead code to disable the bias by applying the disable values of both the pullup and pulldown entries in the table. However, this code is inside the table-lookup loop, so it would only execute if there is an entry for PIN_CONFIG_BIAS_DISABLE in the table, which can never exist, so this code never runs. This commit lifts the offending code out of the loop, so it just executes directly whenever PIN_CONFIG_BIAS_DISABLE is requested, skippipng the table lookup loop. This also introduces a new `param` variable to make the code slightly more readable. This bug seems to have existed when this code was first merged in commit 9dddb4df90d13 ("pinctrl: single: support generic pinconf"). Earlier versions of this patch did have an entry for PIN_CONFIG_BIAS_DISABLE in the lookup table, but that was removed, which is probably how this bug was introduced. Signed-off-by: Matthijs Kooijman Reviewed-by: Haojian Zhuang Reviewed-by: Tony Lindgren Message-ID: <20240319110633.230329-1-matthijs@stdin.nl> Signed-off-by: Linus Walleij

pinctrl: pinctrl-single: Remove some unused fields in struct pcs_function

2024-04-04T19:03:25Z

In "struct pcs_function", the 'pgnames' and 'npgnames' fields are unused. This is a left-over from commit 571aec4df5b7 ("pinctrl: single: Use generic pinmux helpers for managing functions"); Remove them. Found with cppcheck, unusedStructMember. Signed-off-by: Christophe JAILLET Message-ID: Signed-off-by: Linus Walleij

pinctrl: pinctrl-single: move suspend()/resume() callbacks to noirq

2024-03-28T21:05:31Z

The goal is to extend the active period of pinctrl. Some devices may need active pinctrl after suspend() and/or before resume(). So move suspend()/resume() to suspend_noirq()/resume_noirq() in order to have active pinctrl until suspend_noirq() (included), and from resume_noirq() (included). The deprecated API has been removed to use the new one (dev_pm_ops struct). No need to check the pointer returned by dev_get_drvdata(), as platform_set_drvdata() is called during the probe. Reviewed-by: Linus Walleij Reviewed-by: Andy Shevchenko Signed-off-by: Thomas Richard Reviewed-by: Tony Lindgren Reviewed-by: Dhruva Gole Message-ID: <20240102-j7200-pcie-s2r-v4-2-6f1f53390c85@bootlin.com> Signed-off-by: Linus Walleij

pinctrl: pinctrl-single: add ti,j7200-padconf compatible

2023-12-04T14:32:42Z

On j7200, during suspend to ram pinctrl contexts are lost. To save and restore contexts during suspend/resume, the flag PCS_CONTEXT_LOSS_OFF shall be set. Signed-off-by: Thomas Richard Reviewed-by: Tony Lindgren Link: https://lore.kernel.org/r/20231128-j7200-pinctrl-s2r-v1-2-704e7dc24460@bootlin.com Signed-off-by: Linus Walleij

pinctrl: single: Convert to platform remove callback returning void

2023-10-10T13:40:54Z

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 Link: https://lore.kernel.org/r/20231009083856.222030-13-u.kleine-koenig@pengutronix.de Signed-off-by: Linus Walleij

pinctrl: single: Drop if block with always false condition

2023-10-10T13:35:56Z

pcs_remove() is only called after pcs_probe() completed successfully. In this case platform_set_drvdata() was called with a non-NULL argument and so platform_get_drvdata() won't return NULL. Simplify by removing the if block with the always false condition. Signed-off-by: Uwe Kleine-König Link: https://lore.kernel.org/r/20231009083856.222030-3-u.kleine-koenig@pengutronix.de Signed-off-by: Linus Walleij

pinctrl: single: remove get rid of __maybe_unused

2023-09-11T08:49:18Z

These are always used in pcs_probe(). While I was here, I also changed 'unsigned' in the same line to 'unsigned int' to address the checkpatch warnings: WARNING: Prefer 'unsigned int' to bare use of 'unsigned' Signed-off-by: Masahiro Yamada Link: https://lore.kernel.org/r/20230824064508.968142-1-masahiroy@kernel.org Signed-off-by: Linus Walleij

pinctrl: single: Add compatible for ti,am654-padconf

2023-08-10T08:39:17Z

Use the "ti,am654-padconf" compatible to enable the use of wake-up enable and event bits on K3 SOCs that support the daisychain feature Signed-off-by: Tony Lindgren Signed-off-by: Dhruva Gole Link: https://lore.kernel.org/r/20230808102207.130177-3-d-gole@ti.com [Alphabetized the compatible list] Signed-off-by: Linus Walleij

pinctrl: Explicitly include correct DT includes

2023-07-20T19:41:24Z

The DT of_device.h and of_platform.h date back to the separate of_platform_bus_type before it as merged into the regular platform bus. As part of that merge prepping Arm DT support 13 years ago, they "temporarily" include each other. They also include platform_device.h and of.h. As a result, there's a pretty much random mix of those include files used throughout the tree. In order to detangle these headers and replace the implicit includes with struct declarations, users need to explicitly include the correct includes. Signed-off-by: Rob Herring Reviewed-by: Damien Le Moal Acked-by: Emil Renner Berthing Acked-by: Geert Uytterhoeven Link: https://lore.kernel.org/r/20230714174901.4062397-1-robh@kernel.org Signed-off-by: Linus Walleij