Jump to content

"HID Activity" Instant Wake after 2 good sleep/wake cycles (Thunderbolt related? MAG-15)


frickingphil
 Share

4 posts in this topic

Recommended Posts

Hi all,

 

I'm working on this Eluktronics MAG-15 (XMG Fusion / Aftershock Vapor 15 Pro / TongFang QC7) laptop, and I've got pretty much everything working except for a really odd Instant Wake issue. The laptop will sleep and wake perfectly fine without any odd or random wakeups...but only for exactly two cycles (put the laptop to sleep and wake it up), after which it will always instant wake.

 

You can see the difference in the wake log, a "normal" wake says "PWRB XHC/Lid Open" but the broken instant wake lists "Wake from Standby [CDNVA] due to XHC/HID Activity". 

2020-01-27 20:32:51 -0800 Wake  Wake from Standby [CDNVA] due to PWRB XHC/Lid Open: Using AC (Charge:96%) 10 secs   
2020-01-27 20:33:01 -0800 Sleep Entering DarkWake state due to 'Low Power Sleep':TCPKeepAlive=disabled Using AC (Charge:96%)           
2020-01-27 20:33:20 -0800 Wake  Wake from Standby [CDNVA] due to PWRB XHC/Lid Open: Using AC (Charge:96%) 8 secs    
2020-01-27 20:33:28 -0800 Sleep Entering DarkWake state due to 'Low Power Sleep':TCPKeepAlive=disabled Using AC (Charge:96%)           
2020-01-27 20:33:45 -0800 Wake  Wake from Standby [CDNVA] due to XHC/HID Activity: Using AC (Charge:97%)           

It's odd, because this implies it's a USB device that's waking the machine, but there aren't any USB devices attached. I thought it was the trackpad (I2C HID device) but disabling VoodooI2C didn't change anything. 

CLOVER.zip

Edited by frickingphil
Link to comment
Share on other sites

3 hours ago, Hervé said:

If you've not done it, you likely need to patch your USB controller to inject power properties.

 

This may be a stupid question, but this is taken care of when using USBMap to map the ports and generate SSDT-UIAC.aml and SSDT-USBX.aml, correct? If so, then I've already taken care of that...keep in mind the system does successfully sleep (and stay asleep, even overnight) the first 2 times, sleep only breaks and begins instant waking after 2 successful cycles which is really odd.

 

In fact, when I was troubleshooting, I tried disabling all USB ports except for the one the bluetooth module was attached to (for my mouse) and I uninstalled VoodooI2C (to disable the I2C-HID trackpad), and the same problem occurred, except instead of "XHC/HID Activity" as the wake reason, it simply stated "/HID Activity" without the XHC, leading me to think that it's not actually something attached to the USB bus that's causing the wake. 

 

I can't think of any other HID devices attached to the system though...

Link to comment
Share on other sites

27 minutes ago, Hervé said:

What about:

  • BIOS settings (eg: USB wake)?
  • Hibernation settings ?
  • Energy Saver settings?

 

BIOS has wake-on-lan disabled, no options for USB wake (or USB power while asleep). 

 

I've used pmset to set:

  • hibernatemode 0
  • standby 0
  • proximitywake 0
  • ttyskeepawake 0
  • tcpkeepalive 0
  • powernap 0

Wake via Bluetooth is off in bluetooth preferences as well. 

Link to comment
Share on other sites

So I might be onto something...I just realized that most of the other laptops I've seen people have a similar problem with don't have Thunderbolt 3...and this one does.

 

I was messing around with Hackintool and it made me realize that the TB3 port also has its own XHC controller because it also has to support USB-C, so I think this might be the XHC device that's causing the instant wake. 

 

However, disabling Thunderbolt in the BIOS only disables the Thunderbolt side of that port, not the XHC controller...so I'm wondering if there's a way to either:

  • Force the controller to not power on via DSDT / Clover patching?
  • Spoof the PCI ID of the USB-C XHC controller so that AppleXHCITR driver never attaches to it
Link to comment
Share on other sites

 Share

×
×
  • Create New...