[Guide] Installing El Capitan on Spectre x360 (Skylake)

[RehabMan]

Oh... In that post I was referring to the fact that I was using clover's DSDT patches (such as PNLF, Shutdown fix) and addressing how I'll resolve that issue... All the patches will be implemented in my ACPI tables without Clover's auto-patching feature when I'm aware of ALL patches that I need to fix my issues lmao. I.e. I'll address that when I'm done fixing everything. That had nothing to do with me not using another machines ACPI tables. Sorry for the confusion, my wording was a bit off

I would do the patches manually, so you're certain of what you're doing. Clover patching should be used only after you know what works, and then by verifying that Clover is doing what you know to be needed.

 

[ilhanraja]

Of course, I understand. In my last edit of my post I also said that the last attachment that I gave, no clover DSDT patches were being used. I included the PNLF device and the rest of the patches mentioned before are in the patched DSDT that I created. Shutdown fix via Clover previously didn't work and I'm willing to use the one on your GitHub repo (patch not included in the attachment), but like I said before I would like to know what to do beforehand in order to solve my restart/shutdown issues so I can execute the fix as well as have any more ACPI patching issues addressed (don't think there should be any). But thanks for the patience

 

[RehabMan]

Of course, I understand. In my last edit of my post I also said that the last attachment that I gave, no clover DSDT patches were being used. I included the PNLF device and the rest of the patches mentioned before are in the patched DSDT that I created. Shutdown fix via Clover previously didn't work and I'm willing to use the one on your GitHub repo (patch not included in the attachment), but like I said before I would like to know what to do beforehand in order to solve my restart/shutdown issues so I can execute the fix as well as have any more ACPI patching issues addressed (don't think there should be any). But thanks for the patience

Shutdown/restart problems can have many causes...

I would start here:

http://www.tonymacx86.com/threads/guide-native-power-management-for-laptops.175801/

http://www.tonymacx86.com/threads/guide-patching-laptop-dsdt-ssdts.152573/

http://www.tonymacx86.com/threads/guide-10-11-usb-changes-and-solutions.173616/

 

[ilhanraja]

After setting up Clover correctly I have solved the shutdown/restart issues. I won't say ACPI because an issue has arisen on my end.

I went back to this thread because I posted about an issue I was having with specifically HiDPI resolutions... but it seems that the result of the issue is probably incorrectly setup ACPI tables.

Here is a quotation from my original post and the description of the issue.

"On my HP Spectre x360, I have tested many times and every single time the machine boots properly without a display override with any non-HiDPI default resolution in SysPrefs (1920x1080 is default) (actual display is 2560x1440). When I tried adding the display override and changed my display resolution to 1366x768 (HiDPI), I noticed that on some of the boots, it would end up going to second stage and the login (window server) would not appear. There would just be a cursor on the top left (black screen) and that's it.... I would wait for login window to appear and it never would. BUT, on some boots, it would boot right to the login window with my HiDPI resolution. In my window server's plist file DisplayResolutionEnabled is set to YES so it wouldn't be an issue with having HiDPI enabled. But just to be sure, I verified this with Xcode. This never happens when I do not have a display override and am running one of the default resolutions in SysPrefs, the login window would always appear on boot. This is what is perplexing, so this doesn't help me narrow down the issue. I can post a ASL Log and IOReg if needed. NOTE: I have a skylake machine running on 10.11.5"


I have explored the issue further and have tried repatching my ACPI tables and this has not served to help the issue. On certain boots it seems that the window server does not arrive and I am stuck on second stage startup with a cursor. I'll say this happens around 35% of the time.

I have applied the HECI -> IMEI, PNLF, Skylake LPC, Rename GFX0 to IGPU, and the battery status patch available on RehabMan's repos and have removed dynamic SSDT's, Sorted the Order of my ACPI tables loading and have removed all auto-DSDT patches via Clover.

Attached is a screenshot of the kernel log (verbose boot) of the machine before it enters second stage and halts on login when the WindowServer starts as well as my IOReg and CLOVER folder. Nothing looks peculiar as far as what I've setup and nothing seems different in log when the machines starts up normally versus when the machine halts.

 

[RehabMan]

After setting up Clover correctly I have solved the shutdown/restart issues. I won't say ACPI because an issue has arisen on my end.

I went back to this thread because I posted about an issue I was having with specifically HiDPI resolutions... but it seems that the result of the issue is probably incorrectly setup ACPI tables.

Here is a quotation from my original post and the description of the issue.

"On my HP Spectre x360, I have tested many times and every single time the machine boots properly without a display override with any non-HiDPI default resolution in SysPrefs (1920x1080 is default) (actual display is 2560x1440). When I tried adding the display override and changed my display resolution to 1366x768 (HiDPI), I noticed that on some of the boots, it would end up going to second stage and the login (window server) would not appear. There would just be a cursor on the top left (black screen) and that's it.... I would wait for login window to appear and it never would. BUT, on some boots, it would boot right to the login window with my HiDPI resolution. In my window server's plist file DisplayResolutionEnabled is set to YES so it wouldn't be an issue with having HiDPI enabled. But just to be sure, I verified this with Xcode. This never happens when I do not have a display override and am running one of the default resolutions in SysPrefs, the login window would always appear on boot. This is what is perplexing, so this doesn't help me narrow down the issue. I can post a ASL Log and IOReg if needed. NOTE: I have a skylake machine running on 10.11.5"


I have explored the issue further and have tried repatching my ACPI tables and this has not served to help the issue. On certain boots it seems that the window server does not arrive and I am stuck on second stage startup with a cursor. I'll say this happens around 35% of the time.

I have applied the HECI -> IMEI, PNLF, Skylake LPC, Rename GFX0 to IGPU, and the battery status patch available on RehabMan's repos and have removed dynamic SSDT's, Sorted the Order of my ACPI tables loading and have removed all auto-DSDT patches via Clover.

Attached is a screenshot of the kernel log (verbose boot) of the machine before it enters second stage and halts on login when the WindowServer starts as well as my IOReg and CLOVER folder. Nothing looks peculiar as far as what I've setup and nothing seems different in log when the machines starts up normally versus when the machine halts.

Backlight control is not implemented correctly (IntelBacklight.kext not installed).
Not much experience with VoodooHDA...
FakePCIID+FakePCIID_Intel_HD_Graphics might be helpful.
You might try different ig-platform-id values.

 

[ilhanraja]

I've tried IntelBacklight and verified that it was loading... And it didn't resolve the issue.
I tried the FakePCIID_Intel_SKL_Graphics + FakePCIID + 0x1912 device id and that didn't help. I also tried the only other ig-platform-id that was for my GPU 0x1960002 (using 0x1960000) and on some boots it would still lead to the same result... I don't believe this issue is related to VoodooHDA, but I was planning on patching AppleHDA later.

Also, in booterconfig I verified that I had the HiDPI bit mask set (I just turned on all the bits 0xFF)

I haven't verified this to be the exact case, but if I reboot my machine using Windows or via Clover and NOT OS X before I boot, it seems to start up ok.

Maybe my ACPI tables are incorrectly setup? I don't really see what I could've done wrong. Thanks for the suggestions

Would you also like any more logs?

 

[RehabMan]

I've tried IntelBacklight and verified that it was loading... And it didn't resolve the issue.
I tried the FakePCIID_Intel_SKL_Graphics + FakePCIID + 0x1912 device id and that didn't help. I also tried the only other ig-platform-id that was for my GPU 0x1960002 (using 0x1960000) and on some boots it would still lead to the same result... I don't believe this issue is related to VoodooHDA, but I was planning on patching AppleHDA later.

Also, in booterconfig I verified that I had the HiDPI bit mask set (I just turned on all the bits 0xFF)

I haven't verified this to be the exact case, but if I reboot my machine using Windows or via Clover and NOT OS X before I boot, it seems to start up ok.

Maybe my ACPI tables are incorrectly setup? I don't really see what I could've done wrong. Thanks for the suggestions

Would you also like any more logs?

Did you enable legacy boot/CSM?

 

[ilhanraja]

No, since I'm only using UEFI loaders to boot up, I disabled the CSM module/Legacy Support... is that supposed to set to enabled?

 

[RehabMan]

No, since I'm only using UEFI loaders to boot up, I disabled the CSM module/Legacy Support... is that supposed to set to enabled?

Legacy/CSM should be enabled, but still boot UEFI. It solves many graphics issues...

 

[ilhanraja]

Intel HD GPU's generally have a GOP driver available so I presume that it wouldn't be related to that.

But JUST in case, I did that, and right as I started up, the window server didn't arrive... hmm.

It just takes a couple of boots to get it finally loaded correctly