Currently, the screen off/on calls are handled within the suspend sequence, which is a deviation from Windows. This causes issues with certain devices, such as the ROG Ally, which expects this call to be executed with the kernel fully awake. The subsequent half-suspended state makes the controller of the device to fail to suspend properly. This patch calls the screen off/on callbacks before entering the suspend sequence, which fixes this issue. In addition, it opens the possibility of modelling a state such as "Screen Off" that mirrors Windows, as the callbacks will be accessible and validated to work outside of the suspend sequence. Signed-off-by: Antheas Kapenekakis <lkml@xxxxxxxxxxx> --- kernel/power/suspend.c | 16 ++++++++++++++++ 1 file changed, 16 insertions(+) diff --git a/kernel/power/suspend.c b/kernel/power/suspend.c index 19734b297527..afa95271ef00 100644 --- a/kernel/power/suspend.c +++ b/kernel/power/suspend.c @@ -507,6 +507,19 @@ int suspend_devices_and_enter(suspend_state_t state) pm_suspend_target_state = state; + /* + * Linux does not have the concept of a "Screen Off" state, so call + * the platform functions for screen off prior to beginning the suspend + * sequence, mirroring Windows which calls them outside of it as well. + * + * If Linux ever gains a "Screen Off" state, the following callbacks can + * be replaced with a call that checks if we are in "Screen Off", in which + * case they will NOOP and if not call them as a fallback. + */ + error = platform_suspend_screen_off(); + if (error) + goto Screen_on; + if (state == PM_SUSPEND_TO_IDLE) pm_set_suspend_no_platform(); @@ -540,6 +553,9 @@ int suspend_devices_and_enter(suspend_state_t state) Close: platform_resume_end(state); pm_suspend_target_state = PM_SUSPEND_ON; + + Screen_on: + platform_suspend_screen_on(); return error; Recover_platform: -- 2.46.1
