When loading rfkill-gpio driver, the Broadcom RFKILL switches for GPS and Bluetooth are exposed: [ 7545.939852] rfkill_gpio BCM2E1A:00: BCM2E1A:00 device registered. [ 7545.945247] rfkill_gpio BCM4752:00: BCM4752:00 device registered. 0: BCM2E1A:00: Bluetooth Soft blocked: no Hard blocked: no 1: BCM4752:00: GPS Soft blocked: no Hard blocked: no Since the board does not have a Broadcom wireless device on it, I have no idea what these are useful for? Are these GPIO PINs actually exposed somehow? These are platform/80860F0A:00/BCM2E1A:00 and platform/80860F0A:01/BCM4752:00 devices with firmware path \_SB_.URT1.BTH0 and \_SB_.URT2.GPS0 respectively.
can you provide more information on what device this is, how it is connected to the minnowmax?
These switches are exposed via ACPI. On some Baytrail devices they control the reset and shutdown states of Bluetooth and GPS builtin modules. Since the Minnowboard Max does not have these, it have no idea what the behavior would be. So it might be better if these switches are not exposed via ACPI.
Tim, we do not implelment these GPIOs on the minnowmax. they are used to disable wifi/bt/gps via a mPCIe interface. these do not need to be enumerated currently...
When a mini-PCIe Lure is connected, what mechanism do we use to RFKILL?
we don't currently. we did not implement it on either the seacat or flotsam lures. because this mechanism wasn't ironed out at the time of the design, we did not include it. we can always do a revision of the lures as needed... Dave
Hi marcel, These devices have just been removed from our open source release. Please verify it and close this bug.
Created attachment 2234 [details] test logfile Image: core-image-sato image built with poky dizzy release with kernel configured with rfkill_gpio as a module. Firmware: 0.73 open source release as found at https://uefidk.com/content/minnowboard-max
Confirmed fixed as per my test a above. (Sorry for the extra comment - wanted to add an attachment and didn't realize the attachment page was also the comment page)