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 15 Sequoia on Unsupported Macs Thread
- Thread starter flagers
- WikiPost WikiPost
- Start date
- Sort by reaction score
- Status
amaze1499
macrumors 68000
Can we exclude any type of hardware as source of the issue? Different cable, different device, different OS.
Does it shows up in Finder/Locations (any window left hand sidebar)?
Does the device show up when logging into a new user?
Does it show up in ImageCapture?
Can you hear the sound confirming the connection on the iOS device?
Else id run this in terminal
sudo killall PTPCamera
sudo killall imagent
This doesn't exclude OLCP as part of the equation of course.
MacinMan
macrumors 68000
When I was running OCLP, I didn't have any issues on an iMac 17,1 connecting an iPhone 15 Pro Max. Granted that was back with 15.7.2, and an earlier version of iOS 26. Either .0.1, or .1. I know with the new iPad, after upgrading from the stock iPadOS 18 to 26.2, the setup got stuck, and I had to connect it to the iMac to fix it, but it's been fine since. I was running Monterey though, so I can't speak to if it would have worked on OCLP as I had already gone back to supported OS as the host.
Frantisekj
macrumors 65816
Hey, I experienced MBP freeze while asleep. Here is Gemini sumary of crashlog. Not sure it can be helpful for developers. I experience that Mac often wakes fans while in sleep as one fan is louder so I can hear it.
Code:
thought
Here's a thinking process that leads to the suggested analysis of the crash log:
1. **Analyze the User's Input:**
* **Context:** The user provided a crash log after their MacBook Pro (Mid 2012, running macOS 15.7.3 via OCLP) froze and had to be restarted.
* **Question:** "Why did my computer freeze? Here is the log."
* **Log Data:** A standard macOS crash report.
2. **Initial Scan of the Log:**
* **Process:** `Notes` (Apple Notes app).
* **Exception Type:** `EXC_BAD_INSTRUCTION (SIGILL)`. This usually means the CPU tried to execute an instruction it didn't understand, or a software assertion failed.
* **Crashed Thread:** Thread 4 (`com.apple.notes.PaperBundleReader`).
* **System Info:** macOS 15.7.3 (Sequoia), MacBookPro9,1 (Mid 2012 15"), 16GB RAM.
* **Graphics:** Intel HD 4000 + NVIDIA GeForce GT 650M. This is crucial for OCLP on Sequoia.
3. **Deep Dive into the Crash Stack (Thread 4):**
* `0 libswiftCore.dylib ... _assertionFailure ...` -> A Swift assertion failed. This isn't a random memory corruption; the code explicitly checked for something, found it wrong, and crashed the app intentionally to prevent further damage.
* `2 Coherence ...` -> The framework `Coherence` is heavily involved.
* `17 PaperKit ... CRAssetOrData.init(from:) ...` -> `PaperKit` suggests this is related to drawing/handwriting features (Apple Pencil style inputs, even if on a Mac).
* `31 PaperKit ... Image ...` -> Processing an image.
* `58 PaperKit ... CRDataStoreBundle.readPaper(_:url:)` -> Reading a "Paper" bundle (likely a sketch or handwritten note).
* **Interpretation:** The Notes app crashed while trying to read/render a specific attachment (likely a sketch or PDF/"Paper" object) using the `Coherence` and `PaperKit` frameworks.
4. **Connecting to the "Freeze" (The User's Real Problem):**
* The log shows a crash of the *Notes app*, NOT a Kernel Panic (which crashes the whole OS).
* *However*, the user described a full system freeze ("počítač vytuhl... nereagoval").
* *Hypothesis:* The Notes app crash triggered a GPU hang or WindowServer lockup.
* *Evidence in Log:* Look at the loaded libraries.
* `AppleIntelHD4000GraphicsMTLDriver`
* `GeForceMTLDriver`
* *OCLP Context:* macOS Sequoia (15.x) requires Metal 3 for many UI elements (and frameworks like PaperKit). The Nvidia GT 650M (Kepler) and Intel HD 4000 are **Metal 1** (v1.2ish) cards. OCLP patches them to run Sequoia, but advanced rendering features (like those in Notes/PaperKit) often rely on instructions or shaders that these old cards simply cannot handle or emulate perfectly.
5. **Synthesizing the Cause:**
1. The user opened Notes (or it was syncing in the background).
2. Notes tried to render a complex object (Paper/Sketch) using `Coherence`/`PaperKit`.
3. This framework likely used a modern Metal instruction set.
4. The OCLP-patched graphics driver either failed to execute it properly or took too long.
5. Result: The Notes app crashed (SIGILL - Illegal Instruction, often seen when old CPUs/GPUs hit AVX2 or Metal 3 code they don't support).
6. *Side Effect:* This graphics driver struggle likely caused the `WindowServer` (the UI) to hang, making the computer feel "frozen" to the user, leading to the forced restart.
6. **Drafting the Explanation (Czech):**
* **Direct Answer:** The log shows the **Notes app** crashed, specifically when trying to read a "Paper" attachment (sketch/drawing).
* **The "Why":** It's an **AVX2 / Metal compatibility issue** related to OCLP and Sequoia. The `Coherence` and `PaperKit` frameworks (used for new Notes features) use instructions that your 2012 Mac's CPU/GPU doesn't fully support natively.
* **Connection to Freeze:** When the app crashed on a graphics operation, it likely locked up the WindowServer (the interface), making the Mac unresponsive.
* **Specific Error:** `SIGILL` (Illegal Instruction) confirms the CPU/GPU hit a wall.
7. **Actionable Advice:**
* **Immediate:** Avoid the specific note that causes this (if known).
* **Workaround:** Don't use heavy "Sketch" features in Notes on this Mac.
* **OCLP Specifics:** Ensure OCLP and Root Patches are up to date (User already did this, but good to reiterate).
* **Downgrade hint:** Remind that Sequoia on 2012 hardware is experimental for exactly these reasons (Metal 1 vs Metal 3 apps).
8. **Refining the Output (Self-Correction during drafting):**
* *Correction:* I initially thought about RAM, but the log shows `EXC_BAD_INSTRUCTION` inside Swift Core/Coherence. This is definitely software/instruction set mismatch, not hardware failure (like the SSD cable issue from earlier).
* *Tone:* Reassuring. It's not a broken disk this time. It's just software being too new for the hardware.
9. **Final Polish:** Structure with "Co se stalo" (What happened), "Proč" (Why), and "Co s tím" (Solution). (Matches the provided output).analysis:
MacinMan
macrumors 68000
This could be used for development, but it's also a good summary of what happened and why. I've started using Gemini more instead of coming to these forums for help. It's a lot better than asking another person sometimes because it has a lot more information available, and it's also good at giving specifics to my circumstances. Sure, it may need to be corrected once and a while, but I've found, that when I direct its attention to the details missed, it gets it right the second time. I even started an AI Pro subscription because I find the features so useful. Plus AI Pro was the only thing available that combined the two areas I needed features into one.
As far as development, I've found Gemini useful for that as well. It helped me get a simple program written in FutureBasic running. I am used to older languages from the classic mac, and apple days, and even Dos. So Gemini has been a good tool to use to help me combine the old and the new.
josehill
macrumors regular
Frantisekj
macrumors 65816
I know Ai is not perfect but it at least understand it a bit compared to me.
It frozen today again but because of different reason.
Code:
panic(cpu 2 caller 0xffffff8028adef48): PRT1::setPowerState(0xffffff9517307800 : 0xffffff802949e5ac, 0 -> 2) timed out after 30319 ms @IOServicePM.cpp:5609
Panicked task 0xffffff951730d980: 223 threads: pid 0: kernel_task
Backtrace (CPU 2), panicked thread: 0xffffff951fd960c8, Frame : Return Address
0xffffffb062d4fa00 : 0xffffff80283d6761 mach_kernel : _handle_debugger_trap + 0x4c1
0xffffffb062d4fa50 : 0xffffff802854c376 mach_kernel : _kdp_i386_trap + 0x126
0xffffffb062d4fa90 : 0xffffff802853b4cc mach_kernel : _kernel_trap + 0x75c
0xffffffb062d4fb50 : 0xffffff802836b971 mach_kernel : _return_from_trap + 0xc1
0xffffffb062d4fb70 : 0xffffff80283d6a57 mach_kernel : _DebuggerTrapWithState + 0x67
0xffffffb062d4fc70 : 0xffffff80283d60db mach_kernel : _panic_trap_to_debugger + 0x1fb
0xffffffb062d4fce0 : 0xffffff8028bd091d mach_kernel : _panic + 0x81
0xffffffb062d4fdd0 : 0xffffff8028adef48 mach_kernel : __ZN9IOService12ackTimerTickEv + 0x658
0xffffffb062d4fe30 : 0xffffff8028ade8b9 mach_kernel : __ZN9IOService21actionAckTimerExpiredEP8OSObjectPvS2_S2_S2_ + 0x9
0xffffffb062d4fe40 : 0xffffff8028af9e2e mach_kernel : __ZN10IOWorkLoop9runActionEPFiP8OSObjectPvS2_S2_S2_ES1_S2_S2_S2_S2_ + 0x3e
0xffffffb062d4fe80 : 0xffffff8028add498 mach_kernel : __ZN9IOService17ack_timer_expiredEPvS0_ + 0x38
0xffffffb062d4fea0 : 0xffffff8028435b28 mach_kernel : _thread_call_delayed_timer + 0x508
0xffffffb062d4fee0 : 0xffffff8028436b78 mach_kernel : _thread_call_delayed_timer + 0x1558
0xffffffb062d4ffa0 : 0xffffff802836b19e mach_kernel : _call_continuation + 0x2e
Process name corresponding to current thread (0xffffff951fd960c8): kernel_task
Boot args: keepsyms=1 debug=0x100 -lilubetaall -btlfxallowanyaddr ipc_control_port_options=0 -nokcmismatchpanic
Mac OS version:
24G419
Kernel version:
Darwin Kernel Version 24.6.0: Wed Nov 5 21:30:23 PST 2025; root:xnu-11417.140.69.705.2~1/RELEASE_X86_64
Kernel UUID: CF5BCFF5-37A8-32C6-96D4-F6FDC871BAB8
roots installed: 0
KernelCache slide: 0x0000000028000000
KernelCache base: 0xffffff8028200000
Kernel slide: 0x00000000280e8000
Kernel text base: 0xffffff80282e8000
__HIB text base: 0xffffff8028100000
System model name: MacBookPro9,1 (Mac-4B7AC7E43945597E)
System shutdown begun: NO
Panic diags file available: YES (0x0)
Hibernation exit count: 0
System uptime in nanoseconds: 15512721490296
Last Sleep: absolute base_tsc base_nano
Uptime : 0x00000e1bd641167e
Sleep : 0x00000e14bb1037f9 0x000000002ee08ee6 0x00000e08717606cf
Wake : 0x00000e14c62dc4a0 0x000000002edf47f3 0x00000e14c5059821
Compressor Info: 5% of compressed pages limit (OK) and 4% of segments limit (OK) with 2 swapfiles and OK swap space
Zone info:
Zone map: 0xffffff8049a38000 - 0xffffffa049a38000
. PGZ : 0xffffff8049a38000 - 0xffffff804ba39000
. VM : 0xffffff804ba39000 - 0xffffff8518238000
. RO : 0xffffff8518238000 - 0xffffff86b1a42000
. GEN0 : 0xffffff86b1a42000 - 0xffffff8b7e241000
. GEN1 : 0xffffff8b7e241000 - 0xffffff904aa40000
. GEN2 : 0xffffff904aa40000 - 0xffffff951723f000
. GEN3 : 0xffffff951723f000 - 0xffffff99e3a3e000
. DATA : 0xffffff99e3a3e000 - 0xffffffa049a38000
Metadata: 0xffffffe70b05f000 - 0xffffffe72b05f000
Bitmaps : 0xffffffe72b05f000 - 0xffffffe72e05f000
Extra : 0 - 0
straycat23
macrumors member
I have a 2012 Mac Pro that has a Mac Edition Sapphire HD 7950 metal capable graphics card. I know this has probably been addressed before, but I upgraded to OS 15.7.3, which I actually could not upgrade to on this Mac Pro. I have not had any issues with Sequoia 15.7.2 using OCLP, but now I am having issues. Did Apple do something to OS 15.7.3 that would hurt Mac Pro 2012's? I am able to upgrade on my Mac Pro 2013 without problems. The only thing now I can load to my 2012 is Mojave. In other words, I cannot use OCLP or DosDudes Catalina. All my OLCP installers and my DosDude Catalina installer say they are damaged. What can I do to undo all this? Thanks in advance.
Bigwaff
Contributor
How about some specifics regarding your issues? Impossible to offer advice without them.
straycat23
macrumors member
I thought I was being specific. I cannot load either Catalina using Dosdudes upgrade or Big Sur or Sonoma using OCLP. Installer said installation was broken. This was after trying to load Sequoia 15.7.3 from 15.7.2, which worked fine until I tried to upgrade. I’ve been using OCLP for a long time. Not as if I were a novice. I was able to upgrade to 15.7.3 on my Mac Pro 2013, but not my 2012. Installation quits at the end of the upgrade. Says Installer is broken. I noticed OCLP no longer offers Sequoia as an installer. I need help. I usually can figure things out on my own
Bigwaff
Contributor
straycat23
macrumors member
I am using v2.4.1. However, I upgraded to 15.7.3 OTA. It didn’t show as an installer on my machine.
Bigwaff
Contributor
Networking is WiFi or ethernet? Not clear to me whether you successfully updated to 15.7.3 or not. If not, recommend ethernet and create 15.7.3 USB install media using OCLP. Update using USB install media.
straycat23
macrumors member
My network is ethernet. I did not successfully update to 15.7.3. It would not load. Said at the very end that update was damaged and quit. I did discover that the 15.7.3 created a new volume called "update", which I cannot get rid of or change. It curiously reports the same capacity as my HD. It have 3 HD's and an SSD. I've tried to load 15.7.3 on all of them with the same results. I even subsequently bought a new SSD from an eBay vendor with 15.7.3 on it. It would not load. I suspect someone may have put new firmware or a virus on this update that apparently only affects Mac Pro 5.1's. I don't know why they would do that, maybe without intent, but I cannot use OCLP or DosDude's Catalina anymore on my 5.1. I did load Mojave successfully. The guy I bought the SSD with 15.7.3 from said others with 5.1's have reported the same problem. He didn't go into detail. I have loaded successfully 15.7.3 on my Mac Pro 6.1. I will not be doing updates on it until I know it is safe.
Bigwaff
Contributor
Unlikely. Newer versions of macOS do not include MP5,1 firmware updates. Boot Mojave and share your Boot ROM version. I assume it be 144.0.0.0.0 as you already had Sequoia OCLP installed w/ Metal GPU. When updating Sequoia, you need to hang around for reboots. I've found it necessary to hold down Option key on every reboot, select the OCLP EFI in the macOS Boot Picker, then select the appropriate disk in OCLP Boot Picker, typically named "macOS Installer" or something, until the installer no longer appears as a choice.
rehkram
macrumors 65816
In my experience it's been common for new versions to appear for OTA update but not available if you go look for them using the OCLP GUI. They are typically not there for selection for some period of time. Usually they show up within a few days. So this could be a timing issue.
I know this because I prefer to download them to ~/Applications using OCLP and then running the installer manually as a second step. OTA has been glitchy on the rMBP10,1 here for a while now - it hangs between the download/prep step and the restart. So I would prefer to do it manually using OCLP
I know this because I prefer to download them to ~/Applications using OCLP and then running the installer manually as a second step. OTA has been glitchy on the rMBP10,1 here for a while now - it hangs between the download/prep step and the restart. So I would prefer to do it manually using OCLP
Last edited:
Frantisekj
macrumors 65816
I probably realized what was cause. I installed caddy instead of DVD drive. First noname SSD have not worked at all, second Kingstone did but just for while and caused this power management crash. Now I put older 120 GB Samsung there and it works and seems solved crash. I turned off putting SSD into sleep so has to try to turn it on again yet.
straycat23
macrumors member
I always hang around watching the whole update process. I think maybe the key is erasing thoroughly, meanly "XXXX", all my HD's and my one SSD to get rid of this "Update" volume, reinstall Mojave. I have a copy of OLCP on thumb drive as OLCP does not download on Mojave, then build to Sonoma. I don't think I'll risk installing Sequoia again. I'll have to find a way to "XXXX" out my drives, because to my knowledge, the current disc utility no longer allows thorough erasure like it used to.
Bigwaff
Contributor
It will. Use Firefox.
Nuclear option will certainly get rid of it. Do you happen to have older software called "Mountain" installed? There is bug/incompatibility which will mount/display hidden volumes, like "Update" volume.
straycat23
macrumors member
I do not have “Mountain”; however, I think I can dismount the “Update” volume through disc utility, but that will not get rid of it. I used to have a Mac disc eraser, but I think that was meant for IBM processors i.e. before Intel. Were you thinking I had Mountain and that was causing the problem?
saxfun
macrumors regular
two computers running sequoia with OCLP:
1. MacBookPro 2015, 15.7.3 with OCLP 2.4.1 - flawless
2. iMac 13,1 from 2012, upgraded to 16GB RAM and a i7 3770-S CPU. All is running well except the graphic artifacts. I don't think it's the GPU but the version of Sequoia, 15.7.4 DP with OCLP 2.5.0 Nightly. But it was all the same with 15.7.3 and OCLP 2.4.1
Please see attached picture.
Do you know a fix?
1. MacBookPro 2015, 15.7.3 with OCLP 2.4.1 - flawless
2. iMac 13,1 from 2012, upgraded to 16GB RAM and a i7 3770-S CPU. All is running well except the graphic artifacts. I don't think it's the GPU but the version of Sequoia, 15.7.4 DP with OCLP 2.5.0 Nightly. But it was all the same with 15.7.3 and OCLP 2.4.1
Please see attached picture.
Do you know a fix?
Attachments
Sven G
macrumors 6502a
I just discovered that the OCLP 2.5.0 nightly seems to have fixed this well-known Swift runtime error:
github.com
… with now a potential fix here (but obviously not necessary, if it now works in 2.5.0):
… and finally with only this generic warning now present in brew doctor:
Very good…
Doctor: "Warning: No Cask quarantine support available: unknown reason" · Homebrew · Discussion #5482
Output of brew config HOMEBREW_VERSION: 4.3.9 ORIGIN: https://github.com/Homebrew/brew HEAD: 1c98a33a376e5b35ebc801d0f7d6ba7b9eb118ac Last commit: 3 days ago Core tap JSON: 10 Jul 16:17 UTC Core ca...
github.com
… with now a potential fix here (but obviously not necessary, if it now works in 2.5.0):
… and finally with only this generic warning now present in brew doctor:
Very good…
Last edited:
K two
macrumors 68020
Bigwaff
Contributor
Do you see graphic artifacts when booting supported version of macOS? Perhaps the GPU itself is beginning to fail.