pinctrl: amd: Clear GPIO debounce for suspend

soc-button-array hardcodes a debounce value by means of gpio_keys
which uses pinctrl-amd as a backend to program debounce for a GPIO.

This hardcoded value doesn't match what the firmware intended to be
programmed in _AEI. The hardcoded debounce leads to problems waking
from suspend. There isn't appetite to conditionalize the behavior in
soc-button-array or gpio-keys so clear it when the system suspends to
avoid problems with being able to resume.

Cc: Dmitry Torokhov <[email protected]>
Cc: Hans de Goede <[email protected]>
Fixes: 5c4fa2a ("Input: soc_button_array - debounce the buttons")
Link: https://lore.kernel.org/linux-input/mkgtrb5gt7miyg6kvqdlbu4nj3elym6ijudobpdi26gp4xxay5@rsa6ytrjvj2q/
Link: https://lore.kernel.org/linux-input/[email protected]/
Signed-off-by: Mario Limonciello <[email protected]>
Reviewed-by: Hans de Goede <[email protected]>
Link: https://lore.kernel.org/[email protected]
Signed-off-by: Linus Walleij <[email protected]>

Author: superm1

drivers/pinctrl/pinctrl-amd.c

@@ -979,6 +979,17 @@ static int amd_gpio_suspend_hibernate_common(struct device *dev, bool is_suspend
 				  pin, is_suspend ? "suspend" : "hibernate");
 		}
 
+		/*
+		 * debounce enabled over suspend has shown issues with a GPIO
+		 * being unable to wake the system, as we're only interested in
+		 * the actual wakeup event, clear it.
+		 */
+		if (gpio_dev->saved_regs[i] & (DB_CNTRl_MASK << DB_CNTRL_OFF)) {
+			amd_gpio_set_debounce(gpio_dev, pin, 0);
+			pm_pr_dbg("Clearing debounce for GPIO #%d during %s.\n",
+				  pin, is_suspend ? "suspend" : "hibernate");
+		}
+
 		raw_spin_unlock_irqrestore(&gpio_dev->lock, flags);
 	}