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.
MP 1,1-5,1 Mac OS 11.3 has broken support for older Mac Pros
- Thread starter VitaminK
- WikiPost WikiPost
- Start date
- Sort by reaction score
- Status
- Status
I've actually spent the past year (on and off, getting pulled away for other projects) developing a complete AVX/AVX2/BMI1/BMI2/F16C emulator for MacOS. The emulation itself was finished late last year; I've spent the past few months working on the integration (what a surprise - MacOS isn't extremely friendly to hardware spoofing). Single-threaded, it works like a charm, and is faster than I expected. Add a second CPU and/or thread, though, and things go to hell in a hurry. Still, I'm hoping to have something production-worthy in the coming weeks/months...
I don't believe so, 11.4 has been stable as hick only with booting annoyances. we could break Big Sur's snapshot seal if comes down to that...
This is not the first roadblock we had, I hope that is not the last one.
At this moment we really don't know what is the real issue, the crashes are not really helping, but it's clearly PCIe related and the NVMe/AHCI crashes are a consequence, not a cause.
Whether "dubious" or not here's proof. 11.3 RC is booting reliably using HighPoint 7101A-1.
Attachments
The issue being discussed here is MP5,1 and PCIe drives with 11.3/11.4b1, from your signature you are showing a MP3,1, no? I'm failing to see why this exactly helps us since the MP3,1 platform is completely different, the NVMe crash is a consequence, not a cause, and we already have multiple reports of other types of PCIe drives also not working with a MP5,1 and 11.3/11.4b1.
I think so. In the following linked post he is a bit more specific about the used system and wrote clearly about his 3,1… So it’s just another dubious proof, because it’s created on a wrong machine.
https://forums.macrumors.com/thread...unsupported-macs-thread.2242172/post-29807829
Last edited:
I don't doubt that understanding why some older hackintosh configs based with IvyBridge-E/Broadwell CPUs and the MP3,1 can boot 11.3/11.4b1 could be indirectly useful to us, the problem here is that since each MP3,1 user modifies the BootROM itself without any consistency adding APFSJumpStart and NVMe EFI modules from different sources, it's a hard job to replicate it scientifically. Let's not even talk about the hacks.
Since MP61.0120.B00 brought initial NVMe support to MP6,1, Apple released more than 30 firmware updates to it and the user could also injected a MP5,1 module or even from a totally different Mac. If I remember correctly, the APFSJumpStart injected by APFS ROM Patcher is from 2010 unibody MacBooks. Without a dump to understand how the BootROM was made and replicate it, it's just a wild goose chase at this moment.
Ok so tried installing BS11.4 on an SSD on my second DVD bay slot.
I had the same NVMe panics.
Not until I removed all NVMe cards I was able to install it.
Then after reboot it came once with a prohibitory sign. The next 3-4 reboots it was freezing at different stages and on the 5 reboot it booted to desktop.
I still have my
I had the same NVMe panics.
Not until I removed all NVMe cards I was able to install it.
Then after reboot it came once with a prohibitory sign. The next 3-4 reboots it was freezing at different stages and on the 5 reboot it booted to desktop.
I still have my
Sonnet Tempo SSD Pro Plus connected to slot 4, but the SATA SSD's are recognized as PCI devices not PCI-Express.I can get you a bootrom dump and have offered to do so for startergo in another thread. The problem is that the three available tools for dumping the bootrom (ROMtool, Darwin Dumper and Macshrauber) all use the DirectHW.kext, which is no longer supported by 11.3 RC and if an attempt is made to use these tools, the installed kext prevents the Mac booting at the next restart. I suppose the solution is to install another boot disk with High Sierra or Mojave and run ROMtool, etc., from there. If you REALLY DO need this bootrom code to solve the problem, I'll do this in the coming week, but I first have to get another NVMe SSD and build a High Sierra or Mojave USB installer, etc. Let me know.
The ROMTool is working in BS11.4. You have to reboot once after you allow extension. SIP must be disabled as always with the ROMtool.
Last edited:
Sorry but not with mine—yes, I have seen those screens too but reboot stalls. This OCLP 0.0.0.22 install is with SIP disabled but after I allow ROMtool extension install in Sys Prefs—Security & Privacy and reboot, the boot stalls about 1/2 way and the cMP3,1 is "stuck". I then have to use rm -ref in Safe Boot mode to remove the kext and then I can boot again.
Last edited:
Guys we might have a solution. Please try and report. So far so good for me. Next step will be to transfer the installation to NVMe:
github.com
Mac Pro 11.3 Beta NVME issues · Issue #110 · dortania/OpenCore-Legacy-Patcher
Describe the bug Im not sure this will fall within the scope of this patcher, Apple's changes in IONvmefamily has broken support with Mac Pros 5,1/4,1/3,1 the bug started with Beta 3 of 11.3 has co...
github.com
Understandable as there have actually been no reports, dubious or otherwise, from anyone claiming success with booting from a PCIe drive on a 5,1.
this is interesting. I purchased the OWC Upgrade Service for my MacPro 6,1 trashcan late last year. It's been a brick since. It is on its 3rd and last RMA. Seems with High Sierra tech support doesn't run into any issues, but when Big Sur is installed via a clean install, the 6,1 becomes so unstable that it's unusable. most times, it will just keep cycling, trying to boot up, get 20-50% into a boot up then completely go dark. Then it will light up on its own and try to boot again.
I told OWC that if this is confirmed, they need to make it transparent that their Upgrade Service for either the chip or SSD drive, will break the OSX upgrade path, even if Apple has 11.x compatible with 6,1 it doesn't seem to be compatible with their hardware used in the upgrade service.
It was only this past Friday when I realized they were testing on High Sierra and I was installing Big Sur, OWC cleared the machine as fixed and I would get it, install Big Sur and it would be a paperweight. So I told them to test on Big Sur to see if the issue I reported occurs again.
I'll update this post when I hear back from OWC this coming week.
Notice how it says: Aura Pro X2 requires High Sierra OR later.
I told OWC that if this is confirmed, they need to make it transparent that their Upgrade Service for either the chip or SSD drive, will break the OSX upgrade path, even if Apple has 11.x compatible with 6,1 it doesn't seem to be compatible with their hardware used in the upgrade service.
It was only this past Friday when I realized they were testing on High Sierra and I was installing Big Sur, OWC cleared the machine as fixed and I would get it, install Big Sur and it would be a paperweight. So I told them to test on Big Sur to see if the issue I reported occurs again.
I'll update this post when I hear back from OWC this coming week.
Notice how it says: Aura Pro X2 requires High Sierra OR later.
Ooof. Since I’m staring at 11.4 running from an NVME I’ll do this tomorrow. IOPCIFamily really seems the likely cause…Guys we might have a solution. Please try and report. So far so good for me. Next step will be to transfer the installation to NVMe:
Mac Pro 11.3 Beta NVME issues · Issue #110 · dortania/OpenCore-Legacy-Patcher
Describe the bug Im not sure this will fall within the scope of this patcher, Apple's changes in IONvmefamily has broken support with Mac Pros 5,1/4,1/3,1 the bug started with Beta 3 of 11.3 has co...
For those in this thread who have "trust issues" with what other members write about PCIe booting (on a cMP 3,1):
it depends also on the kext you use. I included a kext what run on Big Sur if approved. Also this kext got Dosdude's Romtool to work in BS if copied into ROMTool Application Support folder. This kext is also included into my package.
Last edited:
Unfortunately even after replacing both kexts I still have ionvme panic if my expansion board with the OWC PCIE is attached. Also it hangs if ethernet is connected. This tells me that Ionvme is just a red Herring. It needs to be tested by more people though.
What can we do with this patch Could we replace the IO NVME to be able to use 11.3 or 11.4
- Status