[Guide] Dell XPS 9560 Mojave VirtualSMC, I2C Trackpad, Clover UEFI Hotpatch

[blazinsmokey]

"The developer actually sets the default interval to 160 minutes"
You are right.But VoltageShift was not made intended for Hackintosh.It was made for real macs
Hackintosh´s do not support hibernate.
Next time you put your Hack to sleep and wake it then check with ./voltageshift info
Values returns to -0

VoltageShift is literally a modified version of AnVMSR which was created for the hackintosh community.

hackintosh does support hibernate, rehabman even said so in this thread a while ago.

I think you want to research how writing to MSR registers work to understand why it doesn't matter if it's a real Mac or a hackintosh. The settings stick even after sleep, thats if your laptop sleeps correctly and doesn't hibernate.

 

[ferchizzle]

Btw, I can see you're still having BrcmFirmwareRepo.kext in your clover folder.
This is wrong, I could NOT pair XPS with my iPhone until I replaced it with BrcmFirmwareData.kext
BrcmFirmwareRepo.kext is for L/E or S/L/E ONLY

@blazinsmokey @dmitry_matora is it still necessary to use BrcmFirmwareRepo.kext? I didn’t see that kext in the latest Clover folder that was uploaded. What WiFi/BT cards are you both using?

 

[arehep]

The settings stick even after sleep

False
I just checked it on my HP Elitebook 840 G5 to and its the same.
If i use 30 instead of 1 the values return to -0 after wake from sleep.
From my XPS 9350
Before sleep

After wake from sleep

 

[blazinsmokey]

Hi @blazinsmokey .

Using the new CPUFriend kext does the same thing as the one I generated with this handy script:

GitHub - stevezhengshiqi/one-key-cpufriend: Generate customized CPUFriendDataProvider.kext for dynamic macOS CPU power management

Generate customized CPUFriendDataProvider.kext for dynamic macOS CPU power management - GitHub - stevezhengshiqi/one-key-cpufriend: Generate customized CPUFriendDataProvider.kext for dynamic macOS ...

github.com

Seems less power hungry and battery decreases much slower, but for an unknown reason sleep wake from lid open doesn't work anymore, I have to do it manually via power button. (like you apparently).

Last, I'm sad to hear for your Bios trouble (a bit guilty here)... Maybe try to install new bios 1.15.0 upon it to reset to factory setting https://www.dell.com/support/home/u...N&oscode=WT64A&productcode=xps-15-9560-laptop (after waiting a few days, because even if published apparently dell support aren't even aware it exist " lol We do not have any new BIOS firmware released for your system. We suggest that you do not proceed with the new update.

Hope it'll be ok.

Cheers.

No worries about the BIOS, I knew the risk but now I'm afraid I'm even less help then before lol. I actually updated to 1.15 in hopes but I check HWP bit and its the same so I assume the other ones I changed are as well. I probably take the battery and CMOS battery out this week sometime to reset.

So if it's happening with your device maybe it's not my BIOS settings. I believe what is happening is the laptop doesn't go to sleep at all. It only goes into a deep idle state. After running pmset -g log|grep -e " Sleep " -e " Wake " I get only wake from deep idle and no longer get wake from sleep.

Edit: that's why you have to press power button after opening the lid, it's basically just to wake the screen.

2019-07-05 14:45:07 -0400 Wake                    Wake from Deep Idle [CDNVA] due to PBTN/UserActivity Assertion: Using BATT (Charge:100%)

If you remove the deepidle SSDT, it will sleep but when it wakes the power usage is almost like before but less. Still hovers around 16-17 vs the 22 with previous. Was headed to vacation when this came up so I haven't had a chance to do much other testing.

@blazinsmokey @dmitry_matora is it still necessary to use BrcmFirmwareRepo.kext? I didn’t see that kext in the latest Clover folder that was uploaded. What WiFi/BT cards are you both using?

No I think theres an explanation of this in the first few pages and using the DW1560 as per guide.

 

[blazinsmokey]

False
I just checked it on my HP Elitebook 840 G5 to and its the same.
If i use 30 instead of 1 the values return to -0 after wake from sleep.
From my XPS 9350
Before sleep
View attachment 415306
After wake from sleep
View attachment 415307

In that case I don't think our computers are sleeping properly even if it says it is. From a hardware point of view there is no reason for the MSR to lose changes when going to sleep. Can you try something for me? Disable the launch daemon and set it manually, restart your computer and see if the settings still apply? Our XPS 9560 still holds the offsets after a reboot, even if booting into Windows. The MSR is pretty damn persistent until it's completely powered down. Maybe we are missing some low C-states or something. Also could you check coconutbattery to see if your wattage usage is higher after sleep. Maybe we can figure something out.

 

[arehep]

I removed the launch daemom and shut down.Turned it on and applied -110 -40 -110 .
After reboot these values where still there.
According Coconut it was discharging with 8.09 watts.
Then I put it to sleep for 2 minutes, woke it and values returned to -0
Coconut went up to 8.85

I dont have Windows installed only 10.14.5.
Do you think that's the reason ?

Anyway I returned to sudo ./voltageshift buildlaunchd -110 -40 -110 0 0 0 1 and there's no noticeable change in discharging and if, it´s an improve. 7.74 Watts.
I think this methode is the best way as it also works after shut down - power on

 

[blazinsmokey]

I removed the launch daemom and shut down.Turned it on and applied -110 -40 -110 .
After reboot these values where still there.
According Coconut it was discharging with 8.09 watts.
Then I put it to sleep for 2 minutes, woke it and values returned to -0
Coconut went up to 8.85

I dont have Windows installed only 10.14.5.
Do you think that's the reason ?

Anyway I returned to sudo ./voltageshift buildlaunchd -110 -40 -110 0 0 0 1 and there's no noticeable change in discharging and if, it´s an improve. 7.74 Watts.
I think this methode is the best way as it also works after shut down - power on

Well that's good to hear. If it still holds the offsets after a reboot, something is going wrong in our sleep that's resetting the MSR registers. They should hold through sleep. This was tested on your 9350?

I don't think having a Windows installation matters.

coconut takes a while to retrieve the info and it's not real time. So you may need to idle longer to see lower usage. I would think if it is your 6560U it takes less watts than that. I got our laptops discharging as low as 9 watts and even lower when waking from sleep(deep idle). 7700HQ and 16GB of ram I would think would require more power but I dunno details completely.

 

[arehep]

This was tested on your 9350?

Yes.16gb ram and QHD+ (3200x1800)
Its the QHD+ screen that draw much more power than a FHD.
All tre time I checked with coconut just after Istast Menu finished calculating remaining battery time.
Battery design capacity is 7200 mAh but full charge capacity is only 6317 but it runs about 6 hours fully
charged

The only thing that worry's me is that coconut tels me that battery temperature is -273,2 degrees celsius
Are they made of liquid nitrogen ?

BTW. where's the location of your VoltageShift folder/kext ?
Do you think it should be placed in S/L/E or L/E ?

 

[blazinsmokey]

Yes.16gb ram and QHD+ (3200x1800)
Its the QHD+ screen that draw much more power than a FHD.
All tre time I checked with coconut just after Istast Menu finished calculating remaining battery time.
Battery design capacity is 7200 mAh but full charge capacity is only 6317 but it runs about 6 hours fully
charged

The only thing that worry's me is that coconut tels me that battery temperature is -273,2 degrees celsius
Are they made of liquid nitrogen ?

BTW. where's the location of your VoltageShift folder/kext ?
Do you think it should be placed in S/L/E or L/E ?

We must have the same battery according to coconut. lol. I get same temp reads as well. When working on battery when I first got the laptop I could easily get 8 hours on macOS with it. I've been primarily using it docked now so I can't give you a definitive answer. Last time was a few months ago and I had no problems getting 6 hours with maybe 20-30% to spare. Was only light web browsing and note taking, nothing heavy.

I just have voltageshift in a folder in my downloads folder. Definitely don't put it in SLE or LE. It's in no way a kernel extension. It doesn't really matter where it is b/c after it builds your launch daemon plist which is installed here /Library/LaunchDaemons/com.sicreative.VoltageShift.plist, macOS launchD runs it accordingly.

When you use the command to remove it, it basically just deletes the plist. You can see the code here https://github.com/sicreative/Volta...55bb300d4e50/VoltageShiftAnVMSR/main.mm#L1140

Main reason I just set it manually on boot is because I prefer not to have addition daemons run. I'm stupidly particular about cleanliness, which a lot of the time is pointless I guess. I just open up terminal and 2 bash commands and it's set. One alias to get me to the working directory and another to set it plus others I made to keep things short.

In case anyone is interested in that

# Voltage Shift
alias vs='cd /Users/blazinsmokey/Downloads/voltageshift'
alias vsinfo='./voltageshift info'
alias vsset='./voltageshift offset -125 -90 -125'
alias vsreset='./voltageshift offset 0 0 0'

 

[arehep]

Ok mine is placed at the root of system HD.
I also have a fetisch for keeping the system as vanilla as possible and therefore all kexts for my laptop is placed in Clover/Kexts/Other except BrcmFirmwareRepo.kext and BrcmFirmwareRepo.kext in S/L to get Bluetooth working.
Despite Rehabman´s advice it´s working great.
(Never needed to generate PR file)
But VoltageShift Daemon with my settings never caused any problems at all.
VirtualSmc battery plugin do not show battery cycles on my laptop.
What about your´s ?