Contribute
Register

[HOW TO] OpenCore 0.6.8 > 0.6.9 differences

Joined
Dec 10, 2010
Messages
736
Motherboard
Gigabyte Z390 Aorus Elite
CPU
i9-9900K
Graphics
RX 580
Mobile Phone
  1. iOS
As always, it is highly recommended to read vit9696 message. It is the best place to find out what is new and changed in the new version of OpenCore.

This step from 0.6.8 to 0.6.9 brings quite a few changes and some of them are very important but they are not as many as in the step from 0.6.7 to 0.6.8. Some of the things that have been improved:
  • AppleEvents and some of its keys
  • compatibility with older systems
  • OpenCanopy and the picker presentation
  • OpenCore password decryption speed, especially on slow systems
  • memory devices on MacPro7,1 SMBIOS.
config.plist
  • Misc > Boot > PickerAttributes: OC_ATTR_USE_MINIMAL_UI (0x0040) will hide Restart and ShutDown buttons in the picker (text and OpenCanopy).
  • Misc > Security > ExposeSensitiveData: OCS_EXPOSE_VERSION_UI (0x04) will show OpenCore version in OpenCanopy.
  • Misc > Tools: added support for case-insensitive argument handling in the UEFI tools.

  • UEFI > AppleInput > CustomDelays new type Boolean (former String). It can be False.
  • UEFI > AppleInput > KeyInitialDelay and KeySubsequentDelay failsafe values to Apple OEM values (50 and 1).
  • UEFI > AppleInput > PointerSpeedMul failsafe value changed to Apple OEM value (1).

  • UEFI > Quirks > added EnableVectorAcceleration: enable AVX vector acceleration of SHA-512 and SHA-384 hashing algorithms. Failsafe value is False. If it's true, the signature verification process when OpenCore has a password is faster.

  • UEFI > Quirks > added quirks to increase compatibility with older Mac firmwares:
    - ForgeUefiSupport (Boolean): to workaround legacy EFI 1.x firmwares compatibility. Failsafe value is False.
    - ReloadOptionRoms (Boolean): to force-load Option ROMs on PCI devices. Failsafe value is False.

MacPro7,1 SMBIOS

PlatformInfo > Memory > Update SampleCustom.plist with MacPro7,1 memory samples.

Only for empty slots:
  • PartNumber and SerialNumber now are empty (previous value was 0)
  • Size and Speed = 0 (previous value was 1)
  • Manufacturer = NO DIMM (previous value was empty).
For all slots (filled and empty): DeviceLocator values are ChannelA-DIMM0, ChannelA-DIMM1, ChannelB-DIMM0, ChannelB-DIMM1 and so on up to 12 devices.

It is recommended to use RestrictEvents.kext to eliminate the warning displayed on the desktop at startup.

OpenCanopy
  • Added PickerAttributes: OC_ATTR_USE_MINIMAL_UI to allow running pickers with no Shutdown and Restart buttons.
  • Added ExposeSensitiveData: OCS_EXPOSE_VERSION_UI (0x04) to display OpenCore's version number to OpenCanopy as well as builtin picker.
  • Fallback to builtin picker when OpenCanopy fails to launch or has low resolution.
  • Fixed wraparound when using arrow keys in OpenCanopy: arrow keys will wrap around the boot entry list.
Updated kexts
  • AppleALC.kext
  • IntelMausi.kext
  • Lilu.kext
  • NVMeFix.kext
  • RestrictEvents.kext
  • SMCSuperIO.kext
  • VirtualSMC.kext
  • WhateverGreen.kext.
These kexts have new versions, it's advisable to use them with OpenCore 0.6.9.

How to check config.plist: A guide to update OpenCore 0.x.y to 0.x.y+1
by @Nodarkthings.
 
Last edited:
Joined
Oct 10, 2015
Messages
886
Motherboard
Z370 AORUS Gaming 3
CPU
i5-9600K
Graphics
RX 580
Mac
  1. MacBook
  2. MacBook Pro
Mobile Phone
  1. iOS
The message also refers to a TRIM implementation list that you can check if your SSD works well with TRIM for macOS. It seems SSDs from Western Digital are generally the safest, in terms of NVMe.


Quote from the page for SetApfsTrimTimeout:
  1. 999 (minimal timeout, means TRIM is disabled)
  2. -1 (standard timeout, equals roughly 10 seconds, means TRIM is enabled and runs up 10 seconds during boot)
  3. 4294967295 (maximum timeout, TRIM is enabled and runs as long as needed)

Time consumption of -1 & 4294967295 should be nearly identical when your SSD works well with TRIM.
 
Joined
Jul 11, 2012
Messages
450
Motherboard
Gigabyte GA Z68XP-UD3
CPU
i7 3770
Graphics
GT 210
Mac
  1. iMac
  2. Mac mini
Mobile Phone
  1. Android
Hi! Thought it would be an easy one (as 0.6.7>0.6.8, for me) but hell no. :| Nearly 2mn black screen before picker, boots slower, lightly slower MacOS and had a crash when launching (accidentally) an unsupported PPC app (the same app with 0.6.8 only gives the usual "unsupported" message).

I've replaced the usual files (EFI/BOOT/BOOTx64.efi, EFI/OC/OpenCore.efi, EFI/OC/Drivers/OpenRuntime and EFI/OC/Tools/OpenShell.efi) and updated the values given by OC Config Compare:
Capture d’écran 2021-05-04 à 01.18.29.jpg

My config.plist passes the ocvalidate test too.
Note that my DeviceProperties are for the NVidia GC + headless IGPU. I've deleted the nvda_drv key, just in case, but same result.

Wasn't able to reset NVRAM: Cmd+Opt+P+R didn't work and attempting to display auxiliaries with the space bar froze the picker...
Any idea?
 

Attachments

  • opencore-2021-05-04-090950.txt
    256 KB · Views: 95
  • config.plist
    34.8 KB · Views: 83
Joined
Dec 10, 2010
Messages
736
Motherboard
Gigabyte Z390 Aorus Elite
CPU
i9-9900K
Graphics
RX 580
Mobile Phone
  1. iOS
The message also refers to a TRIM implementation list that you can check if your SSD works well with TRIM for macOS. It seems SSDs from Western Digital are generally the safest, in terms of NVMe...
Thank you for remembering this information. I use 2 Western Digital WD Black NVMe and both seem to work fine with short boot times.
 
Joined
Dec 10, 2010
Messages
736
Motherboard
Gigabyte Z390 Aorus Elite
CPU
i9-9900K
Graphics
RX 580
Mobile Phone
  1. iOS
Hi! Thought it would be an easy one (as 0.6.7>0.6.8, for me) but hell no. :| Nearly 2mn black screen before picker, boots slower, lightly slower MacOS and had a crash when launching (accidentally) an unsupported PPC app (the same app with 0.6.8 only gives the usual "unsupported" message).

I've replaced the usual files (EFI/BOOT/BOOTx64.efi, EFI/OC/OpenCore.efi, EFI/OC/Drivers/OpenRuntime and EFI/OC/Tools/OpenShell.efi) and updated the values given by OC Config Compare...
Yes, your config.plist seems to be well formed. Upload here your entire EFI folder.
 
Joined
Jul 11, 2012
Messages
450
Motherboard
Gigabyte GA Z68XP-UD3
CPU
i7 3770
Graphics
GT 210
Mac
  1. iMac
  2. Mac mini
Mobile Phone
  1. Android
Yes, your config.plist seems to be well formed. Upload here your entire EFI folder.
A bit heavy because I've left every kexts — I'm using old things such as FakeSMC.kext, GenericUSBXHCI.kext and CalDigitUSBxHCI.kext after trying VirtualSMC.kext and a generated USBMap.kext that worked ok, but less suited my needs (less sensors. And I can't get USB2 to work on my USB3 ports with a patched USBMap).

Anyway the issue seems to be at the early stage of OC's loading, as the picker appears only after 2mn — once the OS has launched, everything works (my impression of "lightly slower" was an illusion due to the fact that the Apple screen progress bar is indeed a bit slower, I get the same Geekbench as with 0.6.8; and the crash might also be unrelated).
 

Attachments

  • EFI.zip
    3.7 MB · Views: 80
Joined
Jul 11, 2012
Messages
450
Motherboard
Gigabyte GA Z68XP-UD3
CPU
i7 3770
Graphics
GT 210
Mac
  1. iMac
  2. Mac mini
Mobile Phone
  1. Android
Ah! Talked too fast... As soon as I finished writing, I closed my User account and it crashed!

Code:
*** Panic Report ***
panic(cpu 2 caller 0xffffff8025c8d217): "I/O Kit: Unrecoverable deadlock."@/Library/Caches/com.apple.xbs/Sources/xnu/xnu-3248.73.11/iokit/Kernel/IOService.cpp:1257
Backtrace (CPU 2), Frame : Return Address
0xffffff9171a7b920 : 0xffffff80256dab52 mach_kernel : _panic + 0xe2
0xffffff9171a7b9a0 : 0xffffff8025c8d217 mach_kernel : __ZN9IOService18lockForArbitrationEb + 0x177
0xffffff9171a7b9e0 : 0xffffff8025c90786 mach_kernel : __ZN9IOService5closeEPS_j + 0x26
0xffffff9171a7ba30 : 0xffffff7fa5e4e772 com.apple.iokit.IOStorageFamily : __ZN22IOApplePartitionScheme4scanEPi + 0x44a
0xffffff9171a7bac0 : 0xffffff7fa5e4e2a2 com.apple.iokit.IOStorageFamily : __ZN22IOApplePartitionScheme12requestProbeEj + 0x26
0xffffff9171a7bb00 : 0xffffff7fa5e56130 com.apple.iokit.IOStorageFamily : __ZN7IOMedia11handleCloseEP9IOServicej + 0x1fe
0xffffff9171a7bb50 : 0xffffff8025c907ab mach_kernel : __ZN9IOService5closeEPS_j + 0x4b
0xffffff9171a7bba0 : 0xffffff7fa5e57273 com.apple.iokit.IOStorageFamily : _dkclose + 0x1a4
0xffffff9171a7bbe0 : 0xffffff8025938300 mach_kernel : _spec_close + 0x1e0
0xffffff9171a7bc30 : 0xffffff802593ff9d mach_kernel : _devfs_make_symlink + 0x1b3d
0xffffff9171a7bc80 : 0xffffff8025926757 mach_kernel : _VNOP_CLOSE + 0xd7
0xffffff9171a7bce0 : 0xffffff8025903ef7 mach_kernel : _dounmount + 0x417
0xffffff9171a7bd80 : 0xffffff802590389a mach_kernel : _unmount + 0x19a
0xffffff9171a7bf60 : 0xffffff8025c28341 mach_kernel : _unix_syscall64 + 0x251
0xffffff9171a7bfb0 : 0xffffff80257ec3f6 mach_kernel : _hndl_unix_scall64 + 0x16
      Kernel Extensions in backtrace:
         com.apple.iokit.IOStorageFamily(2.1)[DC1AAB7C-F417-3238-BB3F-2A5B84D67B90]@0xffffff7fa5e4c000->0xffffff7fa5e72fff

BSD process name corresponding to current thread: umount
Boot args: -gux_nosleep keepsyms=1

Mac OS version:
15G22010
 
Joined
Jul 11, 2012
Messages
450
Motherboard
Gigabyte GA Z68XP-UD3
CPU
i7 3770
Graphics
GT 210
Mac
  1. iMac
  2. Mac mini
Mobile Phone
  1. Android
I might have found the issue: copied the EFI folder from my USB key to the ESP of an SSD and it booted fast and all. So, I'm wondering if 0.6.9 is more touchy and doesn't like my USB key? — as it's a 1Mb key, I wasn't able to make Disk Utility create a real ESP (I think it's a limitation of Disk Utility...), I've copied it from the SSD, back in my 0.6.7 trials days... ;-)
Now, wait and see if there's any crash or not.

EDIT: ok, so all is perfect when the EFI is on the SSD. Just cloned it back to a proper ESP on my USB key (starting at sector 40, EF00, etc.) and the looooong black screen came back. So it looks like 0.6.9 doesn't like USB keys...
 

Attachments

  • opencore-2021-05-04-135842.txt
    256 KB · Views: 25
Last edited:
Joined
Dec 10, 2010
Messages
736
Motherboard
Gigabyte Z390 Aorus Elite
CPU
i9-9900K
Graphics
RX 580
Mobile Phone
  1. iOS
I might have found the issue: copied the EFI folder from my USB key to the ESP of an SSD and it booted fast and all. So, I'm wondering if 0.6.9 is more touchy and doesn't like my USB key? — as it's a 1Mb key, I wasn't able to make Disk Utility create a real ESP (I think it's a limitation of Disk Utility...), I've copied it from the SSD, back in my 0.6.7 trials days... ;-)
Now, wait and see if there's any crash or not.

EDIT: ok, so all is perfect when the EFI is on the SSD. Just cloned it back to a proper ESP on my USB key (starting at sector 40, EF00, etc.) and the looooong black screen came back. So it looks like 0.6.9 doesn't like USB keys...
Wait, what size has the USB key? Do you have macOS in a SSD or M2 disk?
 
Joined
Jul 11, 2012
Messages
450
Motherboard
Gigabyte GA Z68XP-UD3
CPU
i7 3770
Graphics
GT 210
Mac
  1. iMac
  2. Mac mini
Mobile Phone
  1. Android
Ok. :lol: I've made a typo! It's a 1Gb USB key.
My main OSes are on an internal SSD.
I made my first OC trials on an external SSD in a USB dock (to be more precise, I've made my really first experiments with it as internal with my three MacOSes in it), then when I saw it worked perfectly I copied the ESP to a 1Gb USB key and it was ok with 0.6.7 and 0.6.8 — tested for about a week before putting it in my internal SSD's ESP to boot my main drive.
 
Top