Haven't tried yet, because stock Catalina IOUSB* worked, but will give a try later.
Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
macOS 10.15 Catalina on Unsupported Macs
- Thread starter 0403979
- WikiPost WikiPost
- Start date
- Sort by reaction score
- Status
Haven't tried yet, because stock Catalina IOUSB* worked, but will give a try later.
Hi @flygbuss
I remember that for Mojave we had to wait for the release of several Beta before having the Product logos.
This time I did them myself... I could not wait... LOL
I partitioned my SSD (APFS) on my supported MBP to install 10.15 since I need the machine for office work and the beta caused too much trouble when I installed it yesterday on my main drive. Everything went well but now, when I'm using 10.14.6 (beta) an additional partition shows up with a - Data suffix.
It's probably already mentioned somewhere but I just I will report it. The drive doesn't show up if I boot 10.15.
Mojave:
Catalina:
Oh and apparently 'iCloud' was moved away from 'Locations'..
It's probably already mentioned somewhere but I just I will report it. The drive doesn't show up if I boot 10.15.
Mojave:
Catalina:
Oh and apparently 'iCloud' was moved away from 'Locations'..
Last edited:
I guess it's the "Users and Library Catalina apfs mount point folder", that typically is hidden to another APFS macOS, but I also noticed when making repartitioning for other macOS installation some hidden partition automatically turn unhidden (and mounted).
Yes it contains a 'Users' and a 'Device' folder.
Right, infact now the real path for "com.apple.boot.plist" is this /Device/Library/Preferences/SystemConfiguration/
apart the other one in the Preboot partition too.
Anyway I have to admit that this new "read-only" Catalina users (and device) folders were really easy to find, I just were confused since I usually create many custom apfs volumes/partitions inside a container.
[doublepost=1559669266][/doublepost]
@ASentientBot you were totally right as always!!!
It's the IOSurface the main responsible that causes kernel panic when trying to preload Nvidia Tesla and IntelHD legacy IOGraphicsFamily for Framebuffer, I just noticed it in my last kernel backtrace.
Please try to fix it!
[doublepost=1559669376][/doublepost]@pkouame a partial good news, I have used the AppleGraphicsControl.kext (with all plugins) and GPUWrangler from HighSierra and they worked and seems also fixed that IOHIDDevice kext_stall , but only for DualGPUs I guess, because on MB7,1 I still have that 60s delay.
The good news is that an HighSierra "Core Framework" has worked on Catalina without any frameworks dependencies pid crashes!!
Last edited:
Booting Catalina on my MBP5,1 always stops at this point:
Any idea? Maybe some IOUSB* replacement?
Any idea? Maybe some IOUSB* replacement?
Most probably, but the fact is that you need first a valid prelinkedkernel to embed them inside it, and it's not simple to create it against the Catalina kernel. But @dosdude1 surely knows how to.
I'd say from my tests was easier since the stock IOUSB* Catalina worked OOB, so I had directly a working keyboard on Catalina single user mode.
Okay, so I've made a little progress, but we're far from done. To sum up:
- A patch to IOHIDFamily.kext _isSingleUser prevents from ever seeing the kextd timeout message
- As @jackluke said, I also independently found that IO80211Family and AppleHDA replacements work
- I nop-ed a couple byte ranges in IOSurface.kext and got (I think) framebuffer acceleration on nVidia Tesla
The bad news: I don't get a crash or anything with the GeForceTeslaGLDriver.bundle. It just... doesn't accelerate.
Anyone else found something interesting?
- A patch to IOHIDFamily.kext _isSingleUser prevents from ever seeing the kextd timeout message
- As @jackluke said, I also independently found that IO80211Family and AppleHDA replacements work
- I nop-ed a couple byte ranges in IOSurface.kext and got (I think) framebuffer acceleration on nVidia Tesla
The bad news: I don't get a crash or anything with the GeForceTeslaGLDriver.bundle. It just... doesn't accelerate.
Anyone else found something interesting?
Great work @ASentientBot !
I've just done another test right now, I only replaced the CoreDisplay.framework but taking from HighSierra and from the Catalina verbose booting I don't encounter any error or pid crashes, rather it seems to flow till the loginUI but then it stucks right after "IOConsoleusers" in verbose lines, a behavior as before your CoreDisplay patch during the 10.14.5 beta, so I've reason to think that we should understand better this new Catalina OpenGL.framework.
Just try replacing only CoreDisplay and you'll notice what I mean, the HighSierra CoreDisplay seems correctly accepted by Catalina kernel and its frameworks but it stays on verbose mode, no stage2 apple logo.
https://forums.macrumors.com/threads/macos-catalina-10-15-bugs-and-bug-fixes.2183833/ you might check out post 13
*** IMPORTANT ***
@dosdude1, @Larsvonhier, @0403979 and everyone who's interested in moving this effort forward: It is important to know that OpenCore is meant as a boot replacement for REAL MACS as well as other types. This is important to note because that is the REAL solution to our problem here that will not only endure, but future proof our efforts.
Please see:
• https://github.com/acidanthera/OpenCorePkg - contains the documentation on compiling and setting up OpenCore.
• https://github.com/acidanthera/OcSupportPkg - support package (OpenCoreShell will be moved into this).
• https://github.com/acidanthera/bugtracker - bug reporting for all of their projects, but also for this (as you can read).
But the thing is will that music app be available for old macos? Like El Capitan for example? Or do you have to be on Catalina for it to happen?
and @ASentientBot . Funny, for a so-called "abandoned' technology, Apple is still modding the OpenGL framework significantly...
What did u NOP in the kext? OpenGL makes a few critical IOSurface calls to get up and running like _IOSurfaceGetID...
BTW. Restoring a TM Backup (Mojave and High Sierra) works flawlessly with Catalina. I made a CCC Backup as well, just in case but I didn't need it.
Reading too many WWDC preview release notes at this point will make you blind...
We will have to see if High Sierra will get the new music app as well as Mojave in a future update but as far as I know they are not abandoning iTunes on Windows yet
#AFAIK OpenCore is still under early development phase and (not really sure) is ready for Release so.. use it with caution.
I don't want to repeat myself but I suppose IOSurface could be related in some way also to the new "Sidecar" Catalina feature.
You could issue the same exact warning about even testing Catalina at this phase. It is obvious that ANYONE in this thread is walking amongst the proverbial dragons.
Nonetheless some people have it working AND the developers need this kind of input and participation at this stage. Obviously no one wants to jump into anything here (Catalina or OpenCore) uneducated or unaware of how to work fluently at the shell level, BUT this should not and cannot be ignored - especially since we're SO EARLY in the beta release cycle with this OS release (Catalina). There's no reason at all to think it's unreasonable to expect OpenCore cannot be ready for prime time within a few months given the rapidity of development that exists (see Github commits).
Oh it is. Same virtual screen tech.
I’m a big fan of any open source project, especially as it relates to the Mac. There is something to be said for cross-pollination and parallel efforts between this project, the hakintosh world and OpenCore. More power to everyone. It’s all good.
That's definitely understandable. I've been knicked before by trying to work with Hackintosh intended stuff (mostly Kexts and EFI) on my Macs
Thats again where the MB4,1 fans get a rest, knowing that accel won‘t happen anyway ;-)
And contrary to some older posts here: 10.14 was / is running a lot better than 10.11 on the GMA x3100 due to changes and optimizations under the hood.
I proved this in various comparison videos and don‘t want to warm it up again. Just mention it here for historians of later generations... ;-)