Search results

I need it confirmed that it wasn't just a config issue all along. If it doesn't boot with the current OC release then I'll clean it up and submit a pull request. @Katanoko1221 are you still using the debug version? Can you upload the log from booting successfully into macOS please.

That's missing the patch I added in this config, https://forum.amd-osx.com/threads/i-need-to-help-with-and-install-issue-with-ryzen-9-9900x.5881/page-4#post-40221 The x86_validate_topology patch.

Yes. Download the official release from GitHub and replace the OpenCore.efi file. You can also change the debug target to 0 as you don’t need to log anything on the release version.

That's excellent news. Next step is to try the official release of OC with that patch added in. All my one is currently doing different is some debug lines and a 0 busratio failsafe.

Strange the patch won't stop OC loading. Copy the patch from the config I uploaded which was just the one fabiosun shared into the last working one you had.

It's possible that the CPU needs the old _x86_validate_topology patch so I've added it specifically for Sequoia. Please test this EFI, Beyond that I'm not too sure why the 9900X is the only CPU that doesn't work.

Can't recall if it was the same in previous logs but OC starts with 14 cores then reads 12 cores :unsure: 00:102 00:013 OCCPU: MP services Pkg 1 Cores 14 Threads 2 - Success 00:412 00:013 OCCPU: Pkg 1 Cores 12 Threads 24 Wondering if that's the reason... Give me some time to look into this.

You'll need a certain one so you still get the BT as it'll need a USB 2.0 header cable on it. If it helps this is the exact one I bought...

That's correct yes.

Out of curiosity can you add these values under Kernel -> Emulate Cpuid1Data = 10 0F 40 0B 00 00 00 00 00 00 00 00 00 00 00 00 Cpuid1Mask = FF FF FF FF 00 00 00 00 00 00 00 00 00 00 00 00

@fabiosun that EFI file is unavailable. If you create an EFI for the user use the last one I uploaded for the debug logs and we'll see where it halts etc.

Well that complicates things lol So it is infact running the correct case, 01:535 00:477 OCCPU: Entered AMD_CPU_EXT_FAMILY_1AH case I'm assuming it just sits on the OpenCore screen with EXITBS etc? It doesn't enter the macOS kernel?

That's how it seems yes. This version doesn't try and force the family but will log which case is being entered. If it's using the correct case for 1Ah then it'll log some extra info to try shed some light on why it's not. If it's showing a family that's not declared then it should be a...

That helps... Let me remove something and I'll upload another test for you tomorrow. It's late here and I'm working early. So it should be running the case for Family 0x1Ah but it wasn't so I was trying to force it by forcing 0x1A which if I'm correct in hex is the same. However it's not...

Hmm if that's correct it's still not running the 1Ah code... Let me add debug strings to all the AMD cases to determine what's being run... Edit: Here we go, each AMD Family case now has a debug string so we know exactly where the 9900x is being routed and will help diagnose why.

It's not the correct build, The one uploaded should have todays date on it. You only need to replace OpenCore.efi and reset the NVRAM.

I see the issue now although it's now being read correctly as Family 1Ah. This is likely caused because now your FSB is not being read so it's trying to divide it by zero. Try this. I've added extra debug lines and a failsafe for the division by zero.

Ok now I see what I was expecting, Does it still not boot?

This seems to be an older build than what I uploaded. Make sure your replacing the OpenCore.efi file The uploaded version should force the correct ExtFamily signature and run the correct code.

Ok, so it wasn't the reason I thought. Try this instead please,