kernel/drivers/input/keyboard/samsung-keypad.c, branch linux-6.18.y

Input: samsung-keypad - use BIT() and GENMASK() where appropriate

2025-07-01T21:31:17Z

Instead of using (1 << ) construct use BIT() helper. Convert (1 << ) - 1 to GENMASK(). Link: https://lore.kernel.org/r/20240819045813.2154642-9-dmitry.torokhov@gmail.com Signed-off-by: Dmitry Torokhov

Input: samsung-keypad - use per-chip parameters

2025-07-01T21:31:10Z

Instead of doing conditional logic based on the chip type, define per-chip parameter structure, and reference it in the match data (either of_device_id or platform_device_id). Reviewed-by: Krzysztof Kozlowski Link: https://lore.kernel.org/r/20240819045813.2154642-8-dmitry.torokhov@gmail.com Signed-off-by: Dmitry Torokhov

Input: samsung-keypad - use guard notation to acquire mutex

2025-07-01T21:31:00Z

Guard notation is more compact and ensures that the mutex will be released when control leaves the function. Reviewed-by: Krzysztof Kozlowski Link: https://lore.kernel.org/r/20240819045813.2154642-7-dmitry.torokhov@gmail.com Signed-off-by: Dmitry Torokhov

Input: samsung-keypad - use devm to disable runtime PM

2025-07-01T21:30:34Z

To make sure that runtime PM is disabled at the right time compared to all other devm-managed resources use devm_pm_runtime_enable(). Link: https://lore.kernel.org/r/20240819045813.2154642-6-dmitry.torokhov@gmail.com Signed-off-by: Dmitry Torokhov

Input: samsung-keypad - use struct_size() helper

2025-07-01T21:16:17Z

When allocating memory for the keypad use struct_size() helper to be protected from overflows. Reviewed-by: Krzysztof Kozlowski Link: https://lore.kernel.org/r/20240819045813.2154642-5-dmitry.torokhov@gmail.com Signed-off-by: Dmitry Torokhov

Input: samsung-keypad - do not combine memory allocation checks

2025-07-01T21:15:58Z

The driver uses devm API to allocate resources so there's no reason to do allocations first and then check and release everything at once. Check results immediately after doing allocation of each resource. Reviewed-by: Krzysztof Kozlowski Link: https://lore.kernel.org/r/20240819045813.2154642-4-dmitry.torokhov@gmail.com Signed-off-by: Dmitry Torokhov

Input: samsung-keypad - do not set input device's parent explicitly

2025-07-01T21:15:45Z

The driver uses devm_input_allocate_device() to allocate instances of input device, which sets the parent appropriately. Remove extraneous assignment. Reviewed-by: Krzysztof Kozlowski Link: https://lore.kernel.org/r/20240819045813.2154642-3-dmitry.torokhov@gmail.com Signed-off-by: Dmitry Torokhov

Input: samsung-keypad - switch to using devm_clk_get_prepared()

2025-07-01T21:15:33Z

Switch to using devm_clk_get_prepared() instead of combining devm_clk_get() with clk_prepare(), which simplifies the code and ensures that the clock is unprepared at the right time relative to releasing other managed resources. Link: https://lore.kernel.org/r/20240819045813.2154642-2-dmitry.torokhov@gmail.com Signed-off-by: Dmitry Torokhov

Input: switch back to struct platform_driver::remove()

2024-10-15T18:43:25Z

After commit 0edb555a65d1 ("platform: Make platform_driver::remove() return void") .remove() is (again) the right callback to implement for platform drivers. Convert all platform drivers below drivers/input/ to use .remove(), with the eventual goal to drop struct platform_driver::remove_new(). As .remove() and .remove_new() have the same prototypes, conversion is done by just changing the structure member name in the driver initializer. While touching these files, make indention of the struct initializer consistent in a few drivers. Signed-off-by: Uwe Kleine-König Link: https://lore.kernel.org/r/20241008090009.462836-2-u.kleine-koenig@baylibre.com Signed-off-by: Dmitry Torokhov

Input: samsung-keypad - convert to platform remove callback returning void

2023-09-24T02:16:41Z

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/20230920125829.1478827-9-u.kleine-koenig@pengutronix.de Signed-off-by: Dmitry Torokhov