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

[InsaneCultist]

On battery or with Adapter? I have this mainly with Adapter, on Battery it is really silent with fan blasting on very rare occasion with 80+ chrome tabs opened.

I have this in both scenarios - I believe I may re-apply the thermal paste on my CPU/GPU today and also do the transistor cooling mod outlined HERE and see how that affects temperatures. However, with the undervolt, the temps are not really a concern. It seems to be cooling the machine adequately, but I cannot seem to get the fan profile to ramp down unless the machine sleeps.

 

[mazakiss]

I have this in both scenarios - I believe I may re-apply the thermal paste on my CPU/GPU today and also do the transistor cooling mod outlined HERE and see how that affects temperatures. However, with the undervolt, the temps are not really a concern. It seems to be cooling the machine adequately, but I cannot seem to get the fan profile to ramp down unless the machine sleeps.

You can do so but don’t expect miracles here. I did this twice and what really improved my temps was mote undervolting than anything else.

Be very careful to only use good and thin 0.5 mm thermal pads for the heatsink part, if not it won’t make proper contact to gpu and chips risking making things worse.

 

[InsaneCultist]

You can do so but don’t expect miracles here. I did this twice and what really improved my temps was mote undervolting than anything else.

Be very careful to only use good and thin 0.5 mm thermal pads for the heatsink part, if not it won’t make proper contact to gpu and chips risking making things worse.

Thanks for the advice. I will perhaps raise my undervolt to 130 from 125 and see how it fairs. My fans may need a cleaning too, I am not sure I have opened this guy up since it was purchased almost 2 years ago.

EDIT: I am testing currently. It appears my fans will ramp down from stage 1 fans (just barely audible) to no fans without issue, but once I really hit the system with something like a 4K video, the fans don't slow down while the video is active, despite 40~ C on the CPU. Even after the video ends, and I am just web browsing, the CPU stays at its low temp but fans stay high. I feel this may be a system issue rather than macOS. I don't really use Windows on this computer all that often, so I am unsure how it fares in Win10.

 

[mazakiss]

If you didn’t open the laptop the last 2 years, cleaning the fans is a good start lol

As far as pasting goes you can do this on the chips above the heatsink as explained.

Particularly in Windows 10 using undervolting fans are even quieter and you can to a certain degree add some fan control even if by default our fans are set up to works at fixed rpm (3 rpm to be exact). You can use intel XTU, or VoltageShift rebooting from osx first boot.

But in osx we can only use these 3 fixed rpm and nothing else even with FakeSMC. (See posts above).

 

[mazakiss]

I don't know if it is useful but I found schematics of XPS 9560. Maybe this can help understand more how things are working.

EDIT: ok I lost my mind, this is certainly copyrighted. sorry we will have to search elsewhere. Sorry mods.

 

[blazinsmokey]

@mazakiss

Appears my last comment and yours was removed by mods... no private communications ok but like just take that part out of my post or something instead of deleting both of our posts which had significant information in it.

I just tested Darkwake=1 as that was originally how KNNSpeed has it set and it works great, no flicker either! Tested sleep and good as well.

You can run this in terminal to help keep track when you're testing.

pmset -g log|grep -e " Sleep  " -e " Wake  "

Give it go, probably better than using darkwake=0

syscl's setup for the 9350 from my understanding didn't have hotplug TB working. I followed many setups and their progressions and his was one of them. I just got a hold of the Alienware 17 R4 files, looks promising. Not hotplug after sleep for TB but could be worth investigating. I can take a look soon. My plans fell through for today so back to the hack.

 

[mazakiss]

Nice!
I also tested different Darkwake values.
Interesting find is that with Darkwake=no you have to disable Handoff in Preferences--->General to get full sleep.
So I test with Darkwake=0, screen flickering was gone but full sleep wasn't working, you are right.
Then used Darkwake=10, same as "no".

What I also noticed is that the pmset -g assertions command result was different I laptop was put into sleep with adapter vs battery.

It's not impossible that VoltageShift can also modify how sleep behave, but I noticed this after uninstalling it to be sure it wasn't related.

Let me try that Darkwake=1 ! I'll report back.

Without being allowed to posting any schematics, you can find via google image the one for the same Alienware 17 R4 used in @au1986 guide. It has the same 7700HQ CPU as in our hack. It show how TB is implemented with PCI lanes etc... Same for XPS9550 which motherboard and TB controller is very close too. Can be instructing maybe.

Thank you @blazinsmokey have a good week.

 

[mazakiss]

I test with Darkwake=1 which leads to the same results as Darkwak=no or 10.

entering terminal command pmset -g log|grep -e " Sleep " -e " Wake " will tell that apparently laptop has full sleep:

2019-05-26 23:50:11 +0200 Sleep                   Entering Sleep state due to 'Clamshell Sleep':TCPKeepAlive=disabled Using Batt (Charge:94%) 25 secs
2019-05-26 23:50:36 +0200 Wake                    Wake from Normal Sleep [CDNVA] due to LID0/HID Activity: Using BATT (Charge:94%) 55 secs
2019-05-26 23:51:31 +0200 Sleep                   Entering Sleep state due to 'Clamshell Sleep':TCPKeepAlive=disabled Using Batt (Charge:94%) 14 secs
2019-05-26 23:51:45 +0200 Wake                    Wake from Normal Sleep [CDNVA] due to LID0/HID Activity: Using BATT (Charge:94%)

But if I test using pmset -g assertions we see that PreventUserIdleSystemSleep is set to 1 when handoff is enabled.

Battery only, Handoff disabled:

Last login: Sun May 26 23:48:46 on console
buzzynesss-macbook-pro:~ buzzyness$ pmset -g assertions
2019-05-26 23:50:56 +0200
Assertion status system-wide:
   BackgroundTask                 0
   ApplePushServiceTask           0
   UserIsActive                   1
   PreventUserIdleDisplaySleep    0
   PreventSystemSleep             0
   ExternalMedia                  0
   PreventUserIdleSystemSleep     0
   NetworkClientActive            0
Listed by owning process:
   pid 98(hidd): [0x0000001c00098024] 00:00:00 UserIsActive named: "com.apple.iohideventsystem.queue.tickle.4294968127.3"
    Timeout will fire in 120 secs Action=TimeoutActionRelease
No kernel assertions.
Idle sleep preventers: IODisplayWrangler

Battery only, Handoff enabled:

buzzynesss-macbook-pro:~ buzzyness$ pmset -g assertions
2019-05-26 23:51:52 +0200
Assertion status system-wide:
   BackgroundTask                 0
   ApplePushServiceTask           0
   UserIsActive                   1
   PreventUserIdleDisplaySleep    0
   PreventSystemSleep             0
   ExternalMedia                  0
   PreventUserIdleSystemSleep     1
   NetworkClientActive            0
Listed by owning process:
   pid 315(sharingd): [0x000000be0001814a] 00:00:01 PreventUserIdleSystemSleep named: "Handoff"
   pid 98(hidd): [0x0000001c00098024] 00:00:00 UserIsActive named: "com.apple.iohideventsystem.queue.tickle.4294968127.3"
    Timeout will fire in 120 secs Action=TimeoutActionRelease
   pid 77(apsd): [0x000000be000b8145] 00:00:01 ApplePushServiceTask named: "com.apple.apsd-datareceived-push.apple.com"
   pid 306(useractivityd): [0x000000be00018143] 00:00:01 PreventUserIdleSystemSleep named: "BTLEAdvertisement"
    Timeout will fire in 58 secs Action=TimeoutActionTurnOff
No kernel assertions.
Idle sleep preventers: IODisplayWrangler

Also depending if you are on battery or adapter, results are different:

Adapter, handoff enabled, now BTLEAdvertisement (blutooth related) is also blocking sleep:

buzzynesss-macbook-pro:~ buzzyness$ pmset -g assertions
2019-05-27 00:01:22 +0200
Assertion status system-wide:
   BackgroundTask                 0
   ApplePushServiceTask           0
   UserIsActive                   1
   PreventUserIdleDisplaySleep    0
   PreventSystemSleep             0
   ExternalMedia                  0
   PreventUserIdleSystemSleep     1
   NetworkClientActive            0
   InternalPreventSleep           1
Listed by owning process:
   pid 315(sharingd): [0x000002d6000181c6] 00:00:18 PreventUserIdleSystemSleep named: "Handoff"
   pid 98(hidd): [0x0000001c00098024] 00:00:00 UserIsActive named: "com.apple.iohideventsystem.queue.tickle.4294968127.3"
    Timeout will fire in 600 secs Action=TimeoutActionRelease
   pid 306(useractivityd): [0x000002e5000181e6] 00:00:02 PreventUserIdleSystemSleep named: "BTLEAdvertisement"
    Timeout will fire in 57 secs Action=TimeoutActionTurnOff
   pid 54(powerd): [0x000002d6000d81bc] 00:00:18 InternalPreventSleep named: "com.apple.powermanagement.acwakelinger"
    Timeout will fire in 26 secs Action=TimeoutActionRelease
No kernel assertions.
Idle sleep preventers: IODisplayWrangler

Adapter, handoff disabled:

buzzynesss-macbook-pro:~ buzzyness$ pmset -g assertions
2019-05-27 00:05:32 +0200
Assertion status system-wide:
   BackgroundTask                 0
   ApplePushServiceTask           0
   UserIsActive                   1
   PreventUserIdleDisplaySleep    0
   PreventSystemSleep             0
   ExternalMedia                  0
   PreventUserIdleSystemSleep     0
   NetworkClientActive            0
   InternalPreventSleep           1
Listed by owning process:
   pid 54(powerd): [0x000003b8000d820b] 00:00:21 InternalPreventSleep named: "com.apple.powermanagement.acwakelinger"
    Timeout will fire in 23 secs Action=TimeoutActionRelease
   pid 98(hidd): [0x0000001c00098024] 00:00:00 UserIsActive named: "com.apple.iohideventsystem.queue.tickle.4294968127.3"
    Timeout will fire in 600 secs Action=TimeoutActionRelease
No kernel assertions.
Idle sleep preventers: IODisplayWrangler

Darkwake=no, or 0 or 1 or 10 give the same results but the process blocking system sleep will be different. Darkwake no, 0 or 1 handoff was blocking system sleep. Darkwake=10, sytem sleep was blocked by evenCalendar.

Really strange. Maybe it's nothing but from what I understand is that display and backligh is sleeping but not system here. Maybe there is a delay here before entering full sleep, I don't know. I retested waiting each time 5 minutes before opening lid and it was showing the same.
attached is my power settings taken from Hackintool.

 

[blazinsmokey]

I test with Darkwake=1 which leads to the same results as Darkwak=no or 10.

entering terminal command pmset -g log|grep -e " Sleep " -e " Wake " will tell that apparently laptop has full sleep:

2019-05-26 23:50:11 +0200 Sleep                   Entering Sleep state due to 'Clamshell Sleep':TCPKeepAlive=disabled Using Batt (Charge:94%) 25 secs
2019-05-26 23:50:36 +0200 Wake                    Wake from Normal Sleep [CDNVA] due to LID0/HID Activity: Using BATT (Charge:94%) 55 secs
2019-05-26 23:51:31 +0200 Sleep                   Entering Sleep state due to 'Clamshell Sleep':TCPKeepAlive=disabled Using Batt (Charge:94%) 14 secs
2019-05-26 23:51:45 +0200 Wake                    Wake from Normal Sleep [CDNVA] due to LID0/HID Activity: Using BATT (Charge:94%)

But if I test using pmset -g assertions we see that PreventUserIdleSystemSleep is set to 1 when handoff is enabled.

Battery only, Handoff disabled:

Last login: Sun May 26 23:48:46 on console
buzzynesss-macbook-pro:~ buzzyness$ pmset -g assertions
2019-05-26 23:50:56 +0200
Assertion status system-wide:
   BackgroundTask                 0
   ApplePushServiceTask           0
   UserIsActive                   1
   PreventUserIdleDisplaySleep    0
   PreventSystemSleep             0
   ExternalMedia                  0
   PreventUserIdleSystemSleep     0
   NetworkClientActive            0
Listed by owning process:
   pid 98(hidd): [0x0000001c00098024] 00:00:00 UserIsActive named: "com.apple.iohideventsystem.queue.tickle.4294968127.3"
    Timeout will fire in 120 secs Action=TimeoutActionRelease
No kernel assertions.
Idle sleep preventers: IODisplayWrangler

Battery only, Handoff enabled:

buzzynesss-macbook-pro:~ buzzyness$ pmset -g assertions
2019-05-26 23:51:52 +0200
Assertion status system-wide:
   BackgroundTask                 0
   ApplePushServiceTask           0
   UserIsActive                   1
   PreventUserIdleDisplaySleep    0
   PreventSystemSleep             0
   ExternalMedia                  0
   PreventUserIdleSystemSleep     1
   NetworkClientActive            0
Listed by owning process:
   pid 315(sharingd): [0x000000be0001814a] 00:00:01 PreventUserIdleSystemSleep named: "Handoff"
   pid 98(hidd): [0x0000001c00098024] 00:00:00 UserIsActive named: "com.apple.iohideventsystem.queue.tickle.4294968127.3"
    Timeout will fire in 120 secs Action=TimeoutActionRelease
   pid 77(apsd): [0x000000be000b8145] 00:00:01 ApplePushServiceTask named: "com.apple.apsd-datareceived-push.apple.com"
   pid 306(useractivityd): [0x000000be00018143] 00:00:01 PreventUserIdleSystemSleep named: "BTLEAdvertisement"
    Timeout will fire in 58 secs Action=TimeoutActionTurnOff
No kernel assertions.
Idle sleep preventers: IODisplayWrangler

Also depending if you are on battery or adapter, results are different:

Adapter, handoff enabled, now BTLEAdvertisement (blutooth related) is also blocking sleep:

buzzynesss-macbook-pro:~ buzzyness$ pmset -g assertions
2019-05-27 00:01:22 +0200
Assertion status system-wide:
   BackgroundTask                 0
   ApplePushServiceTask           0
   UserIsActive                   1
   PreventUserIdleDisplaySleep    0
   PreventSystemSleep             0
   ExternalMedia                  0
   PreventUserIdleSystemSleep     1
   NetworkClientActive            0
   InternalPreventSleep           1
Listed by owning process:
   pid 315(sharingd): [0x000002d6000181c6] 00:00:18 PreventUserIdleSystemSleep named: "Handoff"
   pid 98(hidd): [0x0000001c00098024] 00:00:00 UserIsActive named: "com.apple.iohideventsystem.queue.tickle.4294968127.3"
    Timeout will fire in 600 secs Action=TimeoutActionRelease
   pid 306(useractivityd): [0x000002e5000181e6] 00:00:02 PreventUserIdleSystemSleep named: "BTLEAdvertisement"
    Timeout will fire in 57 secs Action=TimeoutActionTurnOff
   pid 54(powerd): [0x000002d6000d81bc] 00:00:18 InternalPreventSleep named: "com.apple.powermanagement.acwakelinger"
    Timeout will fire in 26 secs Action=TimeoutActionRelease
No kernel assertions.
Idle sleep preventers: IODisplayWrangler

Adapter, handoff disabled:

buzzynesss-macbook-pro:~ buzzyness$ pmset -g assertions
2019-05-27 00:05:32 +0200
Assertion status system-wide:
   BackgroundTask                 0
   ApplePushServiceTask           0
   UserIsActive                   1
   PreventUserIdleDisplaySleep    0
   PreventSystemSleep             0
   ExternalMedia                  0
   PreventUserIdleSystemSleep     0
   NetworkClientActive            0
   InternalPreventSleep           1
Listed by owning process:
   pid 54(powerd): [0x000003b8000d820b] 00:00:21 InternalPreventSleep named: "com.apple.powermanagement.acwakelinger"
    Timeout will fire in 23 secs Action=TimeoutActionRelease
   pid 98(hidd): [0x0000001c00098024] 00:00:00 UserIsActive named: "com.apple.iohideventsystem.queue.tickle.4294968127.3"
    Timeout will fire in 600 secs Action=TimeoutActionRelease
No kernel assertions.
Idle sleep preventers: IODisplayWrangler

Darkwake=no, or 0 or 1 or 10 give the same results but the process blocking system sleep will be different. Darkwake no, 0 or 1 handoff was blocking system sleep. Darkwake=10, sytem sleep was blocked by evenCalendar.

Really strange. Maybe it's nothing but from what I understand is that display and backligh is sleeping but not system here. Maybe there is a delay here before entering full sleep, I don't know. I retested waiting each time 5 minutes before opening lid and it was showing the same.
attached is my power settings taken from Hackintool.

I think it's sleeping fine. Even when I used Darkwake=NO it was sleeping fine. My Assertions look just like yours.

 

[blazinsmokey]

@mazakiss also have you been clearing nvram after darkwake changes?