Clover General discussion

[MakAsrock]

@hnanoto changed .github/workflows/main.yml
 

on:
  push:
    branches:
      - '**'
    tags:
      - '*'
  pull_request:
  workflow_dispatch:

It works better for me this way.
But for some reason he didn't create a pull request. That's a pity. 🙁 

I made a pull request. I hope he won't be upset with me for this.
ci: allow release workflow to run on any tag pattern #1046

Edited January 17 by MakAsrock

[MakAsrock]

Thanks everyone! The latest build, ⌘ Clover revision: 5167 (master, commit 16a4d2cef), feels a bit faster than Open Core 1.0.7 and gets the most out of my Wi-Fi/BT: BCM94360 FENVI FV-HB1200 AC PCI-E adapter.

https://browser.geekbench.com/v6/cpu/16131637
https://browser.geekbench.com/v6/compute/5601048

Edited January 17 by MakAsrock

[Max.1974]

43 minutes ago, MakAsrock said:

Thanks everyone! The latest build, ⌘ Clover revision: 5167 (master, commit 16a4d2cef), feels a bit faster than Open Core 1.0.7 and gets the most out of my Wi-Fi/BT: BCM94360 FENVI FV-HB1200 AC PCI-E adapter.

https://browser.geekbench.com/v6/cpu/16131637
https://browser.geekbench.com/v6/compute/5601048

 

 

That’s right, my friend. There was a restructuring effort in Clover, removing some obsolete code and updating OpenCore. Now we are checking for better integration with OpenCore. A new project coming soon!!  

 

 

Because of this restructuring, I recommend a fresh installation 😉. It’s a different Clover now.
 

[sambul65]

5 hours ago, Max.1974 said:

It’s a different Clover now.

 

Hope someone can make this "different Clover" work without much tinkering with Legacy PCs. 😁 And why would you need to integrate the code with OpenCore? These are competing projects, making them one may lead to quality drop due to lack of competition. 🙃

 

Can you guys add "reconnect -r & map -r" iteration at "different Clover" boot to recognize missed due to slow Legacy PCIe access at 1st pass NVMe drives during the 2nd pass before Clover GUI shows up? That's how earlier releases seems to work before someone broke the code at yet another "restructuring". 😇

 

Edited January 17 by sambul65

[Max.1974]

4 hours ago, sambul65 said:

 

Hope someone can make this "different Clover" work without much tinkering with Legacy PCs. 😁 And why would you need to integrate the code with OpenCore? These are competing projects, making them one may lead to quality drop due to lack of competition. 🙃

 

Can you guys add "reconnect -r & map -r" iteration at "different Clover" boot to recognize missed due to slow Legacy PCIe access at 1st pass NVMe drives during the 2nd pass before Clover GUI shows up? That's how earlier releases seems to work before someone broke the code at yet another "restructuring". 😇

 

 

Clover has had OpenCore embedded in it for quite some time now, right? That is part of the merge. So what is happening? We are improving this new Clover, which already exists. Version 5167 has already been released for end-user use. What happens is that there are projects aimed at advancing technologies and adding new tools to further improve Clover. And Clover has not been competing with OpenCore for a long time now. In fact, one depends on the other. OpenCore by itself is also not just “on its own” — if it were not for the developers of Clover and OpenCore, and vice versa, things would not be the same. So that is it. This is the new Clover that is already there. What will come next, when it comes, will depend on authorization and analysis by the project owners, in this case Slice, and we will keep you all informed here.

 

Now, regarding legacy matters: in my case, I am not using any legacy computer at the moment, but there are specialists here in several channels. There is Chris1111’s GitHub channel, for example, where he has many tools for legacy systems, and that work is being maintained. What I would like to say is this: from now on, if I were you, as I have been doing in my own Clover setups, I copy the keys, copy the keys by equivalence from Clover mirrored in OpenCore, and in this way we bring one closer to the otherI copy them and try to mirror the OpenCore keys into Clover, wherever they are similar or equivalent, as was done in the latest update.

 

They added two keys that exist exactly in OpenCore and exactly in Clover, which means it is now a matter of preference. In the end, it comes down to who keeps the flame burning brighter, those who still keep the Hackintosh vision alive in this scenario.  

Edited January 18 by Max.1974

[deeveedee]

1 hour ago, Max.1974 said:

In the end, it comes down to who keeps the flame burning brighter, those who still keep the Hackintosh vision alive in this scenario.  

 

Well said!

[Max.1974]

@MakAsrock Congratulations, my noble friend, welcome to the team, and thanks to the Administrator. Our Commander-in-Chief, @Allan!! God bless you all!! 

 

[MakAsrock]

2 minutes ago, Max.1974 said:

@MakAsrock Congratulations, my noble friend, welcome to the team, and thanks to the Administrator. Our Commander-in-Chief, @Allan!! God bless you all!! 

 

Hi!

Thanks a lot for the welcome 🙂
I’m excited to be here and help the Hackintosh community as much as I can.
I’ll reach out if I need anything.

Looking forward to working with everyone! 😁

[Max.1974]

3 minutes ago, MakAsrock said:

 

Hi!

Thanks a lot for the welcome 🙂
I’m excited to be here and help the Hackintosh community as much as I can.
I’ll reach out if I need anything.

Looking forward to working with everyone! 😁

 

You are always welcome my friend! 

[sambul65]

On 1/17/2026 at 8:56 PM, Max.1974 said:

I am not using any legacy computer at the moment, but there are specialists here in several channels. There is Chris1111’s GitHub channel, for example, where he has many tools for legacy systems... In the end, it comes down to who keeps the flame burning brighter.

How about offering an "alternative resolution" hack? If YOU code a patch to add "reconnect -r & map -r" iteration at your "different Clover" boot to recognize during the 2nd pass before Clover GUI shows up the NVMe drives missed at 1st pass due to slow Legacy PCIe access, I will test it for you on a Legacy Windows PC. And the 3rd guy will get a "Nobel Price" for "burning the flame brighter". 😇

 

What's important here is the happy end bingo - long lasting peace between Clover fans. 😁 Hope grandma to 8 @deeveedee will approve that, as she certainly has a Legacy PC as well. 🤣

 

Edited January 22 by sambul65

[Max.1974]

@sambul65  Ok my friend, I understand your frustration. I know nostalgia is something amazing in the Hackintosh world — I myself would love to see my old PCs running macOS Snow Leopard, OS X, and so on. But things change, and priorities change too. And even though Clover still has Legacy versions, as I told you, there are pages that still maintain them, bringing NVMe into this era may be incompatible with older systems. And even if it were possible to make newer or semi-new macOS versions work, it would be unfeasible, or drivers, kexts, or hardware simply would not run properly.

 

Still, I believe your idea is a good one, but it needs to be discussed directly with the developers. The coders on the Russian site were questioned about this once, but they said they could not make it happen for a number of reasons, which are not simple — especially when dealing with a bootloader that already emulates the boot process, and even more so with modern versions of macOS.

[sambul65]

13 hours ago, Max.1974 said:

The coders on the Russian site were questioned about this once, but they said they could not make it happen

 

Wow, I wonder how I boot latest Windows from NVMe with Clover on a Legacy PC right now? Its OK however, good luck with your "different Clover" endeavor. 😍

[Max.1974]

5 hours ago, sambul65 said:

 

Wow, I wonder how I boot latest Windows from NVMe with Clover on a Legacy PC right now? Its OK however, good luck with your "different Clover" endeavor. 😍

 

I have no idea how Legacy PC you have... 

My Oldest Hack is this 

 

Update and Working full with Clover 5167

 

 

 

 

Edited January 23 by Max.1974

[Stefanalmare]

Hi! How can these patches be adapted from OpenCore to Clover? I tried but it doesn't work?

CaseySJ-Aquantia-Patch-Sets-1-and-2.plist

Edited January 23 by Stefanalmare

[Max.1974]

6 hours ago, Stefanalmare said:

Hi! How can these patches be adapted from OpenCore to Clover? I tried but it doesn't work?

CaseySJ-Aquantia-Patch-Sets-1-and-2.plist 8.97 kB · 4 downloads

 

@Stefanalmare 

 

EDIT:  I talked with Henrique ( @Hnanoto) and based on the Clover updates, it’s believed that OpenCore plists can be read normally—according to him, you just need to add them. He even updated OpenCore inside Clover precisely for this reason. So maybe it’s enough to copy and paste them into the right places, and Clover may understand the keys normally.

 

There are two places where you can test it: the first would be in KextToPatch and the other in DSDT patches, but you need to convert the numbers to Clover’s format — if I’m not mistaken, Little Endian — and you also need to reduce the number of keys, since many of those from OC are incompatible.

 

As an example converted for Clover:

 

 

You can try too:

 

 

 

 

Edited January 24 by Max.1974

[sambul65]

20 hours ago, Max.1974 said:

I have no idea how Legacy PC you have

 

I built it myself 15 years ago, my Mobo was made in 2008 and has no EFI support, only BIOS. It works perfect for my office needs, and the OS boots normally with Clover USB stick from an NVMe disk hooked to the Mobo PCIe slot as described in that Guide I wrote for 3rd country PC modding fans. I have other hardware as well, modern laptops, but this legacy one supports my 4K monitor and covers basic office tasks. Of course you can't do 4K video editing with it, but I don't anyway. It really runs much faster from NVMe given limited to 8GB RAM size heavily overclocked though. 😉

 

Edited January 24 by sambul65

[Max.1974]

50 minutes ago, sambul65 said:

 

I built it myself 15 years ago, my Mobo was made in 2008 and has no EFI support, only BIOS. It works perfect for my office needs, and boots perfect from Clover USB stick as described in that Guide I wrote for 3rd country PC modding fans. I have other hardware as well, modern laptops, but this legacy one supports my 4K monitor and really covers basic office tasks. Of course you can't do 4K video editing with it, but I don't anyway. It really runs much faster from NVMe given limited to 8GB RAM size heavily overclocked though. 😉

 

 

You are a true enthusiast, really. As for me, on the other hand, I don’t really depend much on good hardware either, but I prefer to stay up to date as a hobby, and I believe it’s easier to deal with desktops. And there are many modern ones at a good price. 

[Avery B]

@sambul65

Opencore was brought in as a dependency due to it's KC (Kernel Collection) injection capabilities. With Big Sur, Apple changed the kernel cache format and did some cleanup in the kernel, breaking kext injection in Clover. As part of that change, the format changed from the Prelinked Kernel to the KC. Am unsure why that would break legacy booting though - Opencore itself is capable of booting on legacy platforms through duet (and has been tested to boot 10.4-26). At a first glance, I don't think the limitations would be on the OC side but maybe I'm wrong.

[Henry2010]

OK my build cannot start after installing 5167 after I added the new quirks to the config.
I installed 5166 again but it would be nice to have some answers.

Since all kexts are of the latest stable version, I assume the old DSDT.aml is broken when using it with 5167?

Thanks.

 

The error messages contain these lines:

Quote

appleacpicpu processorid=31 localapicid=255 disalbed

acpi error method parse/execution failed

 

Edited January 27 by Henry2010

[Matgen84]

3 hours ago, Henry2010 said:

OK my build cannot start after installing 5167 after I added the new quirks to the config.
I installed 5166 again but it would be nice to have some answers.

Since all kexts are of the latest stable version, I assume the old DSDT.aml is broken when using it with 5167?

Thanks.

 

The error messages contain these lines:

 

 

Hi @Henry2010 What is your settings for new quirk, especially disableiomappermapping ?

 

For Alder Lake, according to Opencore guide :

<key>Disableiomapper</key>
<true/>
<key>DisableIoMapperMapping</key>
<true/>

If I understand well,  in order for our patched ACPI Table to load properly we must also drop the current DMAR table:

 

<key>DropTables</key>
<array>
     <dict>
          <key>Signature</key>
          <string>DMAR</string>
     </dict>

For your DSDT patched, you can test patching from Clover GUI, press F5.

[Marcinho]

4 hours ago, Henry2010 said:

OK my build cannot start after installing 5167 after I added the new quirks to the config.
I installed 5166 again but it would be nice to have some answers.

Since all kexts are of the latest stable version, I assume the old DSDT.aml is broken when using it with 5167?

Thanks.

 

The error messages contain these lines:

 

Take a look at the settings "ResizeAppleGpuBars", set to "0".

 

[Henry2010]

@Matgen84 @Marcinho

 

Thank you.

I edited these entries in the config.plist, and tested 5167 again with a flashdrive. It still doesn't work.
I'll stick with 5166 and macOS 15 for another year.

It's just puzzling that this update would break a working config.

[jlrycm]

22 hours ago, Matgen84 said:

 

Hi @Henry2010 What is your settings for new quirk, especially disableiomappermapping ?

 

For Alder Lake, according to Opencore guide :

<key>Disableiomapper</key>
<true/>
<key>DisableIoMapperMapping</key>
<true/>

If I understand well,  in order for our patched ACPI Table to load properly we must also drop the current DMAR table:

 

<key>DropTables</key>
<array>
     <dict>
          <key>Signature</key>
          <string>DMAR</string>
     </dict>

For your DSDT patched, you can test patching from Clover GUI, press F5.

If you are patching and loading a DMAR table with the reserved regions removed, the quirk DisableIOmapper must be set as “False” in the config file and VTD must be enabled in BIOS. If you don’t keep VTD enabled, the patched DMAR solution wouldn’t work.

[Marcinho]

6 hours ago, Henry2010 said:

@Matgen84 @Marcinho

 

Thank you.

I edited these entries in the config.plist, and tested 5167 again with a flashdrive. It still doesn't work.
I'll stick with 5166 and macOS 15 for another year.

It's just puzzling that this update would break a working config.

I had the same error as you when I updated to version 5167. To fix it, I used my old config.plist, only changing the ResizeAppleGpuBars quirk setting from "-1" to "0". I didn't change or add any other settings. Have you tried this method?

Just to note that this was, for me, a very good update; the system boots up faster, is even more fluid and stable, Geekbench results are better, in short, it was worth it.

Best of luck and happy hacking!

Edited January 28 by Marcinho

[Henry2010]

@jlrycm @Marcinho

 

Yes, ResizeAppleGpuBars is set to 0.

Thanks again, but I don't plan to test again because of the soaring prices of SSD and RAM.

 

This config and the EFI folder have been working perfectly with all the Clover updates these years.

A bit confusing to see it becomes incompatible this time.

Edited January 29 by Henry2010