-
Notifications
You must be signed in to change notification settings - Fork 0
Home
These are more detailed install notes as I went through the guide and tried different things to get each feature working.
I did a bios dump for myself for future reference, committed to the repo. It will only be helpful to you if you have the exact same motherboard. Basically, I started with the BIOS recommendations in the Dortania guide:
- Fast Boot
- Secure Boot
- Serial/COM Port
- Parallel Port
- VT-d (can be enabled if you set DisableIoMapper to YES)
- Compatibility Support Module (CSM)
- Intel SGX
- Intel Platform Trust
- CFG Lock
- VT-x (called Intel Virtualization in the BIOS)
- Above 4G Decoding (Leave bar resize disabled)
- Hyper-Threading
- Execute Disable Bit
- EHCI/XHCI Hand-off
- OS Boot Type: UEFI
- DVMT Pre-Allocated(iGPU Memory): 64MB or higher
- SATA Mode: AHCI
Used USBToolkit. Followed this guide. You can map prior to installing Mac OS in windows, or insert USBInjectAll.kext to install, enable XHCIPortlimit, and check things are working before doing the full mapping.
Notes: Macs have a 15 port USB limit and this motherboard has more than 15 ports (USB 3 ports are counted as 2 ports: supporting both USB 3 and USB 2, as well as USB-C 3 ports supporting switch connections when you plug in either way). I disabled some of the USB-C port rotations (so you can only plug in the cable one way, and spinning the wire won't work) to be under the 15-port limit.
I set this to be disabled as a monitor driver, and just used for gfx calculations. I plug monitors directly into dGPU. I used AAPL,ig-platform-id=0300913E
in DP config.
Recommended framebuffers (Desktop):
-
0x3EA50000
(default) -
0x3E9B0007
(recommended) input as07009B3E
- Empty framebuffer (CFL):
0x3E910003
(default) input as0300913E
- Empty framebuffer (CML):
0x9BC80003
(default)
This was such a pain to get working. I ended up using layout alcid=1
even though I don't have an SPDIF/Optical out on this board. It still works though.
For some reason it stopped working while tinkering with something else in the config and then eventually just came back online and working 100% of the time. Randomly sometimes works, sometimes not. A few times I’ve booted and all the onboard audio devices show up. Tried to debug this to get it to work all the time. Went through each layout-id, some caused kernel panics, others booted but didn’t result in working onboard audio being detected.
- tried adding alcdelay=500->3000 as per guide, no change
- tried adding SSDT-HPET.aml to ACPI as per guide, dumped DSDT.aml from OpenCore using Quirk Misc > Debug > SysReport (grab DSDT.aml from thumb drive in SysReport USB, then run https://github.com/corpnewt/SSDTTime on hackintosh to create SSDT-HPET.aml (add to ACPI, then add created patches). No change :/
Realtek ALC887 0x100202, 0x100302, layout 1, 2, 3, 5, 7, 11, 12, 13, 17, 18, 20, 33, 40, 50, 52, 53, 87, 99
Had to add agdpmod=pikera
to boot args
The biggest issue for this motherboard was that MMIO whitelisting was required for consistent booting. Without going through that process, maybe 1 in 3 boots would result in getting stuck at ([EB|#LOG:EXITBS:START])[https://dortania.github.io/OpenCore-Install-Guide/troubleshooting/extended/kernel-issues.html#stuck-on-eb-log-exitbs-start] when booting.
21:479 00:063 OCABC: MMIO devirt start 21:543 00:063 OCABC: MMIO devirt 0xE0000000 (0x10000 pages, 0x800000000000100D) skip 0 (3758096384) 21:606 00:063 OCABC: MMIO devirt 0xFE000000 (0x11 pages, 0x8000000000000001) skip 0 (4261412864) 21:670 00:063 OCABC: MMIO devirt 0xFEC00000 (0x1 pages, 0x800000000000100D) skip 0 (4273995776) 21:734 00:063 OCABC: MMIO devirt 0xFED00000 (0x4 pages, 0x800000000000100D) skip 0 (4275044352) 21:797 00:063 OCABC: MMIO devirt 0xFEE00000 (0x1 pages, 0x8000000000000001) skip 0 (4276092928) 21:861 00:063 OCABC: MMIO devirt 0xFF000000 (0x1000 pages, 0x800000000000100D) skip 0 (4278190080) 21:924 00:063 OCABC: MMIO devirt end, saved 278620 KB 21:988 00:063 OCABC: All slides are usable! You can disable ProvideCustomSlide!
From hackintosh discord bot, slide calc function (memmap.txt exported from opencore boot debug logs)
$slide memmap.txt
Applicable Slide Values:
0x100000: slide=0
0x1000: slide=0
0x091000: slide=0
$slide page # highest
Applicable Slide Values:
0x77E000: slide=4
33:020 00:028 OCABC: MMIO devirt start
33:049 00:028 OCABC: MMIO devirt 0xE0000000 (0x10000 pages, 0x800000000000100D) skip 0 = 3758096384
33:077 00:028 OCABC: MMIO devirt 0xFE000000 (0x11 pages, 0x8000000000000001) skip 0 = 4261412864
33:105 00:028 OCABC: MMIO devirt 0xFEC00000 (0x1 pages, 0x800000000000100D) skip 0 = 4273995776
33:134 00:028 OCABC: MMIO devirt 0xFED00000 (0x4 pages, 0x800000000000100D) skip 0 = 4275044352
33:162 00:028 OCABC: MMIO devirt 0xFEE00000 (0x1 pages, 0x8000000000000001) skip 0 = 4276092928
33:191 00:028 OCABC: MMIO devirt 0xFF000000 (0x1000 pages, 0x800000000000100D) skip 0 = 4278190080
33:219 00:028 OCABC: MMIO devirt end, saved 278620 KB
- DisableIOMapper False
- Booter DevirtualiseMmio True
- Add Booter > MmioWhitelist to each of the MMIO devirt listed in debug printout while booting check which ones provided stable boot and exclude otherr mem addresses
Common issue with hackintoshes. Eventually I found out it was just taking a LONG time (like 10s) to fall asleep and not much was broken. I did have issues with auto wake though, see below. Problem fixed however.
At first I thought sleep wasn’t working. Screens would lose signal but fans would still be running and clearly not sleeping. I tried a couple things like:
- https://dortania.github.io/OpenCore-Post-Install/universal/sleep.html#preparations. Also recommends disabling some wake settings from BIOS but my mobo didn’t have any of those.
- https://dortania.github.io/OpenCore-Post-Install/universal/sleep.html#irq-conflicts. I ran through this, and I think this is what actually fixed my sleep? Not sure. It’s in my config and doesn’t seem to harm anything. Will boot up into old EFI without this patch and check that sleep is working. Edit: sleep works without this patch, just have to wait 10s.
- https://dortania.github.io/OpenCore-Post-Install/universal/sleep.html#audio. It also mentions audio. Randomly my mobo audio started working and sleep is also working so maybe it’s this.
- Add Wake Type Property: https://dortania.github.io/OpenCore-Post-Install/usb/misc/keyboard.html#method-1-add-wake-type-property-recommended (did this): PciRoot(0x0)/Pci(0x14,0x0) was my USB XHC PCIRoot. Doesn’t seem necessary. Also I never had to push anything other than space bar to wake.
I have intermittent issue with sleep, sometimes immediately wakes up. Logs say:
2022-09-30 15:59:21 -0600 Sleep Entering Sleep state due to 'Software Sleep pid=158':TCPKeepAlive=disabled Using AC (Charge:0%) 16 secs
2022-09-30 15:59:37 -0600 DarkWake DarkWake from Normal Sleep [CDN] : due to PEG1 PEGP PEG2 PEGP RP01 PXSX RP02 PXSX RP03 PXSX RP04 PXSX PXSX RP06 PXSX PXSX PXSX RP10 PXSX RP11 PXSX RP12 PXSX RP13 PXSX R/Network Using AC (Charge:0%) 14 secs
2022-09-30 15:59:51 -0600 Wake DarkWake to FullWake from Normal Sleep [CDNVA] : due to UserActivity Assertion Using AC (Charge:0%)
^ Issue is the DarkWake from Normal Sleep [CDN]
Wake for network access wasn’t disabled despite running pmset from command line. Turning that off fixed auto wake.
- seems to be caused by 2nd point of https://dortania.github.io/OpenCore-Post-Install/universal/sleep.html#rtc. Adding Kernel > Quirks > DisableRtcChecksum fixes it!
Still not working in Safari, but everything works in Chrome/Firefox e.g., can watch Netflix, etc. Added shikigva=80
to boot args, did not fix. https://dortania.github.io/OpenCore-Post-Install/universal/drm.html#fixing-drm