On Sat, Dec 23, 2017 at 03:53:57PM +0000, Ghorai, Sukumar wrote:
included in 4.14-rc1, so something needs to be done in Linus's tree to resolve this issue, otherwise people will hit this as a regression when moving to 4.14 or newer.
Well, I wouldn't object to reverting it in Linus' tree too, since AFAIK, this is something that can be configured by user-space policy, no? And that way, no user space is surprised by the sudden additional requirements on managing bluetooth wakeup.
Wakeup from Bluetooth will be observed as long as USB wakeup is configured from user space or kernel space or connected to any Bluetooth HID devices. And we don't want to support the wakeup from BT HID devide. The solution looks like disable Bluetooth in S3 entry? Looks we are trying to hide the other problems in a system, and reverting this patch.
I agree, please work with the bluetooth developers to make this happen. They seem to be ignoring patches right now, so you might have to kick them a few times :(
Do you suggest submitter to send the revert request to 4.12 and 4.14 kernel?
4.12 is long end-of-life, I don't know what you expect to happen there.
4.14 is already released, this needs to be done for 4.15 first, right?
thanks,
greg k-h