Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
From
Mykola Grymalyuk today
about:

[dortania/OpenCore-Legacy-Patcher] macOS Sonoma and OpenCore Legacy Patcher Support (Issue #1076)]

"With the unveiling of macOS 14, Sonoma, Apple has continued their march towards removal of older Intel Macs. And with this, many more critical drivers used by our older systems have been changed or outright removed.

As the OS has been in developer beta for less than a few days, we've compiled a list of challenges the patcher will need to overcome for this fall. This list will likely expand as we determine more issues with older systems.


  • Dropped Models
  • Graphics support
    • Metal
      • 31001
      • 3802
    • Non-Metal
  • Wireless support
  • Bluetooth Support
  • Overall timeline for support

Dropped Models​

Overall a fairly tame OS release, project will need to expand support for the following models:

MacBook10,1: MacBook (Retina, 12-inch, 2017)
MacBookPro14,1: MacBook Pro (13-inch, 2017, Two Thunderbolt 3 ports)
MacBookPro14,2: MacBook Pro (13-inch, 2017, Four Thunderbolt 3 Ports)
MacBookPro14,3: MacBook Pro (15-inch, 2017)
iMac18,1: iMac (21.5-inch, 2017)
iMac18,2: iMac (Retina 4K, 21.5-inch, 2017)
iMac18,3: iMac (Retina 5K, 27-inch, 2017)

Currently the only known limitation with these units is the lack of wireless support. Otherwise graphics is still natively supported.

Graphics support​

31001-based Graphics Cards​

With macOS Sonoma, Apple has transitioned to a new LLVM compiler version known as 32023. This compiler is what powers the Metal graphics stack in macOS, and is critical to getting functional acceleration.

32023

However it seems this new compiler version hasn't changed the ABI drastically, as GPU's based off of the 31001 compiler are still mostly functional. Currently we've determined UI corruption within IOSurface, which would lead to incorrect rendering of wallpapers and menubar.

  • We have mostly resolved this issue internally, however cannot ship due to complications IOSurface ABI.
Default​
Experimental Patch​
Screenshot 2023-06-05 at 7 59 39 PM
Additionally we've discovered an ABI change in IOAcceleratorFamily2 that causes kernel panics when engaging with the Intel framebuffer. Currently investigating the cause, but problematic functions are:

IOAccelResource::pageonIfNeeded()
IOAccelResource::pageoffIfNeeded()

Downgrading IOAcceleratorFamily.kext works around this panic issue, however for long term maintance this needs to be resolved properly.


Graphics cards under 31001 designation:

Intel Broadwell

# Applicable Models:
MacBook8,1
MacBookAir7,x
MacBookPro12,1
iMac16,x
Intel Skylake

# Applicable Models:
MacBook9,1
MacBookPro13,x
iMac17,1
AMD Legacy GCN

# Applicable Models:
MacBookPro11,5 (Note iGPU is not 31001 based, so model is overall broken)
MacBookPro13,3
iMac15,1
iMac17,1
MacPro6,1

3802-based Graphics Cards​

As mentioned in the 31001 section, Apple transitioned to a new LLVM version 32023. This has major implications with 3802 GPUs, as we're currently barely working with the Ventura release.
Specifically we're relying on 13.2.1's 31001 binaries to run, and as such 32023 will not be compatible with these GPUs.

At the time of writing, there has been no success at all with Sonoma, and we believe this will be our biggest challenge for support. Overall we're unsure if we can support these graphics cards.


Graphics cards under 31001 designation:

Intel Ivy Bridge

# Applicable Models:
MacBookAir5,x
MacBookPro9,x
MacBookPro10,x
iMac13,x
Macmini6,x
Intel Haswell

# Applicable Models:
MacBookAir6,x
MacBookPro11,x
iMac14,x
iMac15,1 (internal, headless iGPU)
Macmini7,1
Nvidia Kepler

# Applicable Models:
MacBookPro9,1
MacBookPro10,1
MacBookPro11,3
iMac13,x
iMac14,x

Non-Metal​

As more of macOS Ventura relies on the WindowManager stack introduced with macOS Ventura, no timeline can be given for support. Similar to 3802, we're unsure if we'll be able to get non-Metal functional in Sonoma.


Graphics cards under non-Metal designation:

Intel Ironlake

# Applicable Models:
MacBookPro6,x
Intel Sandy Bridge

# Applicable Models:
MacBookAir4,x
MacBookPro8,x
iMac12,x
Macmini5,x
AMD TeraScale 1 and 2

# Applicable Models:
MacBookPro8,2
iMac7,1
iMac8,1
iMac10,1
iMac11,x
iMac12,x
Macmini5,2
MacPro3,1
MacPro4,1
MacPro5,1
Nvidia Tesla (as well as Nvidia Web Drivers)

# Applicable Models:
MacBook5,x
MacBook6,1
MacBook7,1
MacBookAir2,1
MacBookAir3,x
MacBookPro4,1
MacBookPro5,x
MacBookPro6,x
MacBookPro7,1
iMac7,1
iMac8,1
iMac9,1
iMac10,1
iMac11,x
iMac12,x
Macmini5,2
MacPro3,1
MacPro4,1
Xserve2,1
Xserve3,1

Wireless support​

With Sonoma, another big axe has been dropped: Removal of IO80211FamilyLegacy support

This stack was responsible for supporting the majority of wireless cards used Macs, including machines up-to iMac18,x.

We currently have a highly experimental patch set that restores support, however is currently not feasible for end users due to the reliance on a Kernel Debug Kit for building. Proper solution is still under investigation.

BCM94350 Demo​
BCM94360 Demo​

Wireless cards under IO80211FamilyLegacy designation:

BCM943224, BCM94331, BCM94350, BCM94360, BCM943602

  • Applicable to all unsupported Macs that had native wifi in Ventura

Bluetooth Support​

Currently only a minor issue, however Bluetooth scanning is currently non-functional on all unsupported Macs. Current work-around is to downgrade bluetoothd, proper solution is being investigated.

Overall timeline for support​

As with every new major release of macOS, developing a patcher becomes more and more challenging for our small group of hobbyists. As it stands, we're unsure when we'll be able to resolve all these issues.

Regarding proper support for macOS Sonoma on OpenCore Legacy Patcher, no accurate prediction can be made. But for those who'd wish for a rough estimate, we'd say 6 months from now when we're able to support Sonoma. The challenges presented with 3802, non-Metal and legacy wireless will unfortunately require extensive time and research to develop functional patch sets.

However otherwise, we hope everyone's excited for what's to come!"

I believe they will be able to solve it again :)
 
@Coselmann Great info - thanks for posting that. Disappointing to be sure, but not unexpected. At this point I'd consider a MacPro6,1 being supported under Ventura 13.4 via OCLP 0.6.6 to be nothing short of miraculous. I have very little expectation about a Trash Can ever being able to run Sonoma 14.x.

[Edit: and of course only hours later my "little expectation" has been met. lol]
 
Last edited:
@Coselmann Great info - thanks for posting that. Disappointing to be sure, but not unexpected. At this point I'd consider a MacPro6,1 being supported under Ventura 13.4 via OCLP 0.6.6 to be nothing short of miraculous. I have very little expectations about a Trash Can ever being able to run Sonoma 14.x.
Ventura gave us some time. We shall see.
 
  • Like
Reactions: Riot Nrrrd
At least for the rest of the Macs we can work to resolve the issues with Ventura so at least that OS will be the end for most of our unsupported Intel Macs. I am grateful for at least that. The OpenPatcher has given more life to a few of my old Macs.

Thanks team for the hard work!
 
I am not a developer but I am very much ready to be a Guinea Pig. I have a ‘13 Mac Pro that I am OK to trash - all I really do is on my ‘19 Mac Pro and my ‘22 Mac Studio.

Are you saying it would possibly work to install Sonoma on my MP6,1? Any particular steps - or just do it?
 
I think I would hold off on Sonoma just yet. Even with the night version OCLP. As for DRM, Apple TV effectively freezes my 'trash' so it's not working yet. Netflix ok. No issues.
 
I'm not sure if OCLP correctly works with Ventura with the trash can: mine just crashed all of a sudden and rebooted.

What is more interesting: the fan stopped working after the crash, so I had to reset SMC to make it work again.
 

Attachments

  • MacPro6,1_crash_ventura.txt
    11.2 KB · Views: 85
  • Screenshot 2023-06-16 at 8.33.30 PM.png
    Screenshot 2023-06-16 at 8.33.30 PM.png
    147 KB · Views: 73
I have so far experienced several situations where the system "froze up". I also applied an SMC reset just in case. It appeared that the problem was with the drive (WD_BLACK 2TB SN770 M.2 2280 PCIe Gen4) but in general I don't have any drastic problems like you. It also "froze" the system on my MBP late 2013 twice so maybe there really is still something wrong with Ventura on older Apples. And still sometimes the system ejects my external drives by itself and here I still don't know the cause.
 
And still sometimes the system ejects my external drives by itself and here I still don't know the cause.
Well, unfortunately, that's not even uncommon for supported models.
It's the main reason I reverted back to Big Sur on my 2020 Intel MBP (it was a complete fresh Ventura install).
 
  • Love
Reactions: Coselmann
Oh thank you very much for that. You have saved me time to experiments! I have already tested with 4 NVME drives on my computers with no success . Sometimes you just have to ask :)
 
  • Like
Reactions: arw
I have so far experienced several situations where the system "froze up".
I think they are still running down applications that use AVX2. So, if you recall what you were doing at the time it froze, it would be helpful to report it here so the devs can look into it.
 
  • Like
Reactions: Coselmann
I have a ‘14 MBP and the ‘13 MP, both running Ventura via OCLP. Both working fine. Both no apparent issues.
 
Diagnostics Information (past 7-30 days):
2023-06-18 08:17:18 mediaanalysisd - Crash (24 times)
Executable: /System/Library/PrivateFrameworks/MediaAnalysis.framework/Versions/A/mediaanalysisd
2023-06-16 18:48:55 signpost_reporter - High CPU Use
Executable: /usr/libexec/signpost_reporter
2023-06-15 06:46:28 photoanalysisd - Crash (23 times)
Executable: /System/Library/PrivateFrameworks/PhotoAnalysis.framework/Versions/A/Support/photoanalysisd
2023-06-11 23:55:55 com.apple.WebKit.WebContent - High CPU Use (2 times)
Executable: /Volumes/VOLUME/*/W**************k/V******s/A/X*********s/c*****************************c/C******s/M***S/c*************************t
 
panic(cpu 2 caller 0xffffff800bdec85e): nvme: "3rd party NVMe controller. Loss of MMIO space. Write. fBuiltIn=1 MODEL=WD_BLACK SN770 2TB FW=731100WD CSTS=0xffffffff US[1]=0x0 US[0]=0x9 VID=0x15b7 DID=0x5017 CRITICAL_WARNING=0x0.\n" @IONVMeController.cpp:6147
Panicked task 0xffffff87d3aa4038: 273 threads: pid 0: kernel_task
Backtrace (CPU 2), panicked thread: 0xffffff87d3cc0b30, Frame : Return Address
0xffffffa180d7baa0 : 0xffffff800966f4fd mach_kernel : _handle_debugger_trap + 0x4ad
0xffffffa180d7baf0 : 0xffffff80097c3c54 mach_kernel : _kdp_i386_trap + 0x114
0xffffffa180d7bb30 : 0xffffff80097b36d9 mach_kernel : _kernel_trap + 0x3c9
0xffffffa180d7bb90 : 0xffffff800960f951 mach_kernel : _return_from_trap + 0xc1
0xffffffa180d7bbb0 : 0xffffff800966f7dd mach_kernel : _DebuggerTrapWithState + 0x5d
0xffffffa180d7bca0 : 0xffffff800966ee87 mach_kernel : _panic_trap_to_debugger + 0x1a7
0xffffffa180d7bd00 : 0xffffff8009ddceab mach_kernel : _panic + 0x84
0xffffffa180d7bdf0 : 0xffffff800bdec85e com.apple.iokit.IONVMeFamily : __ZN16IONVMeController18RequestAsyncEventsEj
0xffffffa180d7be20 : 0xffffff8009d1c466 mach_kernel : __ZN18IOTimerEventSource15timeoutSignaledEPvS0_ + 0x96
0xffffffa180d7be70 : 0xffffff8009d1c376 mach_kernel : __ZN18IOTimerEventSource17timeoutAndReleaseEPvS0_ + 0xc6
0xffffffa180d7bea0 : 0xffffff80096c5958 mach_kernel : _thread_call_delayed_timer + 0x508
0xffffffa180d7bee0 : 0xffffff80096c6a08 mach_kernel : _thread_call_delayed_timer + 0x15b8
0xffffffa180d7bfa0 : 0xffffff800960f19e mach_kernel : _call_continuation + 0x2e
Kernel Extensions in backtrace:
com.apple.iokit.IONVMeFamily(2.1)[54A1C402-2099-332C-B960-FF81C8B25619]@0xffffff800bde4000->0xffffff800be10fff
dependency: com.apple.driver.AppleMobileFileIntegrity(1.0.5)[68F4D402-252A-3D76-B1CE-FF5D5BB11D0D]@0xffffff800abde000->0xffffff800ac11fff
dependency: com.apple.iokit.IOPCIFamily(2.9)[6E72A292-C2AA-3F1B-8141-213720748CBD]@0xffffff800c07c000->0xffffff800c0adfff
dependency: com.apple.iokit.IOReportFamily(47)[491DDA55-D371-3A28-9A77-C28DB013D8AE]@0xffffff800c0be000->0xffffff800c0c0fff
dependency: com.apple.iokit.IOStorageFamily(2.1)[D7517DDF-BFD9-3191-94A7-8973E077BA4E]@0xffffff800c1b1000->0xffffff800c1c8fff

Process name corresponding to current thread (0xffffff87d3cc0b30): kernel_task
Boot args: keepsyms=1 debug=0x100 ipc_control_port_options=0 -nokcmismatchpanic

Mac OS version:
22F66

Kernel version:
Darwin Kernel Version 22.5.0: Mon Apr 24 20:51:50 PDT 2023; root:xnu-8796.121.2~5/RELEASE_X86_64
Kernel UUID: 7E997BC9-2104-3D4D-9AAE-17BD7A3FEC2D
roots installed: 0
KernelCache slide: 0x0000000009200000
KernelCache base: 0xffffff8009400000
Kernel slide: 0x00000000092dc000
Kernel text base: 0xffffff80094dc000
__HIB text base: 0xffffff8009300000
System model name: MacPro6,1 (Mac-F60DEB81FF30ACF6)
System shutdown begun: NO
Panic diags file available: YES (0x0)
Hibernation exit count: 0

System uptime in nanoseconds: 110221734491
Last Sleep: absolute base_tsc base_nano
Uptime : 0x00000019a9ba3e02
Sleep : 0x0000000000000000 0x0000000000000000 0x0000000000000000
Wake : 0x0000000000000000 0x00000008b272da5c 0x0000000000000000
Compressor Info: 0% of compressed pages limit (OK) and 0% of segments limit (OK) with 0 swapfiles and OK swap space
Zone info:
Zone map: 0xffffff8160703000 - 0xffffffa160703000
. PGZ : 0xffffff8160703000 - 0xffffff8170704000
. VM : 0xffffff8170704000 - 0xffffff863ad6a000
. RO : 0xffffff863ad6a000 - 0xffffff87d3a36000
. GEN0 : 0xffffff87d3a36000 - 0xffffff8c9e09c000
. GEN1 : 0xffffff8c9e09c000 - 0xffffff9168702000
. GEN2 : 0xffffff9168702000 - 0xffffff9632d68000
. GEN3 : 0xffffff9632d68000 - 0xffffff9afd3cf000
. DATA : 0xffffff9afd3cf000 - 0xffffffa160703000
Metadata: 0xffffffd763224000 - 0xffffffd783224000
Bitmaps : 0xffffffd783224000 - 0xffffffd793224000
Extra : 0 - 0
Fresh from today :(
 
Hey all,

First off, fantastic work with OCLP/MacPro 6,1. I was able to install OCLP and Ventura very easily, patch and spoof without hardly any issue. I have universal control working very well with my work laptop. So hats off to you.

Second if this isn't the right place for this, I'm new here, and happy to take it elsewhere if you point me in the right direction.

The issue I'm running into is MS word. Of course we all wish there was a better alternative, but in my industry MS office is never going to go away and working around those products is simply not worth it. With that out of the way:

I'm finding larger documents with vector graphics, images, tables and references render slowly (10's to 100's of pages). Scrolling is extremely choppy on some docs. My trash can should be more than adequate for this variety of work (12-core CPU/64GB RAM/D500) and yet still it's painful to work in MS word (beyond the normal). The same document imported into Apple Pages scrolls adequately though not amazing (but the formatting is borked from the conversion so Pages isn't a real alternative). Is there anything about the graphical acceleration for word processing that may not be kicking in to enable smooth functioning here? Is this just poor software design + old machine = bad performance?

Anyhow, not sure if there is a best place for this, but this thread seems to deal specifically with keeping the Trash Can alive.

Thanks again!
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.