Hi All, This is a rewrite of a patch which I wrote a while ago: https://www.spinics.net/lists/kernel/msg3474327.html The old version got nacked because it was changing behavior based on whether s2idle or s3 suspend was used at probe-time, while this can be changed at run-time. This new version instead delays calling enable_irq_wake() until the drivers' suspend call-back is called, at which point we know for sure which type of suspend is being used (for this suspend/resume cycle), thus addressing the problem with the old version. Regards, Hans Hans de Goede (1): platform/x86: intel_int0002_vgpio: Only call enable_irq_wake() when using s2idle drivers/platform/x86/Kconfig | 2 +- drivers/platform/x86/intel_int0002_vgpio.c | 80 +++++++++++++++------- 2 files changed, 57 insertions(+), 25 deletions(-) -- 2.31.1
