Clover General discussion

[JorgeMax]

29 minutes ago, Matgen84 said:

Rigth, BOOTX64.efi is a rename file of CLOVERX64.efi. When you update Clover from PKG, both files are changed to the new version.  Post debug.log (Notebook) maybe users can help you.

Hi @Jief_Machak

I update my IvyBridge config.plist to new standard. I can boot r5124 but my MaLdOn's DSDT failed to display PCI entries in System Info (Mojave). All works fine with r5122.

Any ideas ! Please

 

Still, nothing of my DSDT is recognized, I missed some corrections
My processor appears in "on this mac" as unknown, it is an ivy bridge (i5-3570K). Should it be some information conflict inserted from my CPU in the system? Does Clover identify a way and OpenCore inside it other information?
I can't even send the debug log, because it is blank "zero bytes"

[maclinuxG4]

does anyone make a test with R5123 or r5124 on high sierra and Nvidia video board?

thanks

Edited October 13, 2020 by maclinuxG4

[Jief_Machak]

2 hours ago, Matgen84 said:

Rigth, BOOTX64.efi is a rename file of CLOVERX64.efi. When you update Clover from PKG, both files are changed to the new version.  Post debug.log (Notebook) maybe users can help you.

I propose that Clover package install BootloaderChooser instead, to definitively put an end at this duplication of clover efi file.

 

2 hours ago, JorgeMax said:

I can't even send the debug log, because it is blank "zero bytes"

I'm working on that.

2 hours ago, Henry2010 said:

VirtualSMC.efi since 5124 includes OC now. Maybe that means we need to remove FakeSMC* kexts?

VirtualSMC.efi and FakeSMC.kext has nothing to do with each other. One is loaded at EFI time, one is loaded by kernel.

2 hours ago, Matgen84 said:

I can boot r5124 but my MaLdOn's DSDT failed to display PCI entries in System Info (Mojave). All works fine with r5122.

We'll do "differential" debugging together, like before and we'll find the problem.

 

Hey everyone : don't go too fast  I can't keep up... When a bug is known, no need to test anymore until one dev says it's fixed !

[PG7]

2 hours ago, Matgen84 said:

Rigth, BOOTX64.efi est un fichier renommé de CLOVERX64.efi. Lorsque vous mettez à jour Clover à partir de PKG, les deux fichiers sont modifiés vers la nouvelle version. Post debug.log (Notebook) peut-être que les utilisateurs peuvent vous aider.

salut @Jief_Machak

Je mets à jour mon IvyBridge config.plist au nouveau standard. Je peux démarrer r5124 mais le DSDT de mon MaLdOn n'a pas pu afficher les entrées PCI dans System Info (Mojave). Tout fonctionne bien avec r5122.

Des idées ! S'il vous plaît

 

Idem 

 

DSDT no charged

 

HP Envy Reline 23-K030ef 

haswell !!!

 

Clover r5123 = OK

Clover r5124 = no charged DSDT ! 

[iCanaro]

28 minutes ago, Jief_Machak said:

I propose that Clover package install BootloaderChooser instead, to definitively put an end at this duplication of clover efi file.

Spoiler

 

28 minutes ago, Jief_Machak said:

2 hours ago, JorgeMax said:

I can't even send the debug log, because it is blank "zero bytes"

I'm working on that.

 

let me know when you'll be ready to see you do something about AMD, X570 my T-RyZo

 

for now a huge thank you to you and Slice for the excellent work done so far

PS: a mention for Fusion always precise and clarifying interventions

Edited October 13, 2020 by iCanaro

[naiclub]

Now waiting for version 5125 to come out, I will try. In case of correction for DSDT.aml to work

[Mirone]

I installed clover r5124 and cannot boot Catalina or Big Sur with my 5123 EFI folder I can boot Catalina but not Big Sur, I am rusty.

[tikizondo94]

Im having an issue with 5123 and 5124 loading voodooi2c kexts in the correct order. No matter what my laptop will not boot Catalina with these new versions of Clover. 5122 and all before work normal. Is there a way to set the order of kext loading with the new clover? 

[Jief_Machak]

1 hour ago, tikizondo94 said:

kext loading with the new clover? 

Not yet. On the TODO list.

[BuXb]

20 hours ago, fusion71au said:

 

The latest version Clover_r5124_a2ffdf864

                          Clover_r5124_086e5682a.pkg

 

On my legacy desktop GA-P55aUD3 (System 2 in signature)...

  Hide contents

 

Some tips for converting over from old Clover r5122 to new ones with OC 0.6.1 integration in this post.

wow, thank you, fusion71au , for all the great information, screenshots and your edits! Who would've thought.. I was mislead by Clover stopping at 10.11 in kexts/Other.

 

I'll be back at that Laptop in a few days and can report back then. First then need to convert from ole dinosaur Chameleon (remember ?) to Clover. Should be fun.

Edited October 13, 2020 by BuXb

[Vampirexx]

9 hours ago, Matgen84 said:

@Vampirexx First, a lot of Quirks necessary to new Clover are missing in your config.plist. Have a look to sample.plist  What is quirks.efi ? you mean Openruntime.efi, I suppose.

And try again.
 

Hi buddy after last clover and setting demo config from https://www.insanelymac.com/applications/core/interface/file/attachment.php?id=358633

 

now i got this kernel panic

[SavageAUS]

13 hours ago, Henry2010 said:

I strongly recommend using Clover 5124.PKG instead of replacing *.efi files.

PKG does not install on Big Sur. 

[Henry2010]

1 hour ago, SavageAUS said:

PKG does not install on Big Sur. 

 

I meant that for 10.15 and earlier.

But maybe we can make some changes to Clover for 10.16?

 

/usr/local/bin/bdmesg, espfinder, partutil

~/.local/bin/bdmesg, espfinder. partutil

 

/EFI-Backup(s?)

~/.EFI-Backup(s)

Edited October 14, 2020 by Henry2010

[LockDown]

@fusion71au

Im a little bit confuse of the changes of your previous posts. can you enlighten me mate

 

You said:

Quote

However this commit r5123_641b75e by @Slice changes old Clover KernelAndKextPatches behavior:
To get power management for Sandy/IvyBridge CPUs that need patched AppleIntelCPUPowerManagement.kext, now need KernelPm=Yes instead of AppleIntelCPUPM=Yes.
To get kernel power management for Haswell and newer CPUs , now need KernelXCPM=Yes instead of KernelPm=Yes.

 

@Slice replied:

Quote

I understand now. Fixed. See 086e5682a

 

You edited your post with:

Quote

Edit:  Bug fixed in Clover_r5124_086e5682a.pkg
Edited 5 hours ago by fusion71au 
KernelAndKextPatches behavior restored with r5124_086e5682a.

 

Now my question is...

Which settings is correct now starting at Clover_r5124_086e5682a

Sandy/Ivy > KernelPm=Yes
Haswell+ > KernelXCPM=Yes

OR

 

Sandy/Ivy > AppleIntelCPUPM=Yes
Haswell+ > KernelPm=Yes

 

[SavageAUS]

20 hours ago, Matgen84 said:

You use MacBookPro16,1 (i9) instead of MacBookPro15,1 (i7): so you need CPU patch in your config.plist display the correct processor (cosmetic). I think.

Fixed.

I updated smbios to 16,4 which did not change it so i just set cpu-type in clover to 0x0705

[fusion71au]

32 minutes ago, ellaosx said:

@fusion71au

Im a little bit confuse of the changes of your previous posts. can you enlighten me mate

 

You said:

 

@Slice replied:

 

You edited your post with:

 

Now my question is...

Which settings is correct now starting at Clover_r5124_086e5682a

Sandy/Ivy > KernelPm=Yes
Haswell+ > KernelXCPM=Yes

OR

 

Sandy/Ivy > AppleIntelCPUPM=Yes
Haswell+ > KernelPm=Yes

 

 

Old (correct) behavior was restored with Clover_r5124_086e5682a

 

ie

Sandy/Ivy > AppleIntelCPUPM=Yes
Haswell+ > KernelPm=Yes

 

The first scenario was applicable only to r5123_641b75e to r5124_a2ffdf8, due to the bug introduced in r5123_641b75e.

[naiclub]

5 hours ago, mifjpn said:

Sorry for the late reply.
I'm glad that the build is working well on your computer.
For added accuracy, it is recommended to check the build number with the following command.
% cd
% cd CloverBootloader
% git rev-parse --short HEAD

Thanks, I will try

[Slice]

15 hours ago, Henry2010 said:

 

 

No idea what MaLdOn means but maybe you can try generating a new DSDT?

 

Also, I wonder if we still need VirtualSMC.efi since 5124 includes OC now. Maybe that means we need to remove FakeSMC* kexts?

 

I have to explain.

There are two main roads leading to purpose.

One, older way supporting by me and still working:

1. FakeSMC.kext+its plugins (IntelCpuMonitor.kext, RadeonMonitor.kext, ITEIT87x.kext, ACPIMonitor.kext, VoodooBatterySMC.kext, SMIMonitor.kext)

2. SMCHelper.efi driver

Second way, developed by vit9696 and Acidanthera group:

1. VirtualSMC.kext+its plugins (.....). It required Lilu.kext to be also installed.

2. VirtualSMC.efi driver.

 

Don't mix these different ways! Don't jump from one road to another on full speed!

[Ludox]

12 hours ago, Mirone said:

I installed clover r5124 and cannot boot Catalina or Big Sur with my 5123 EFI folder I can boot Catalina but not Big Sur, I am rusty.

 

Hello my dear friend, all right?
Tell me what you use your desktop or notebook, let's see if I can help you this time ..

What is the problem ...?
I have no problem with the clover 5123 and 5124 all perfect Big Sur and Catalina both on the desktop PC and on the Acer e1 notebook in signature.

 

 

[LockDown]

3 hours ago, fusion71au said:

 

Old (correct) behavior was restored with Clover_r5124_086e5682a

 

ie

Sandy/Ivy > AppleIntelCPUPM=Yes
Haswell+ > KernelPm=Yes

 

The first scenario was applicable only to r5123_641b75e to r5124_a2ffdf8, due to the bug introduced in r5123_641b75e.

Thank you for clarification.

That said, , what does kernelXCPM now for?

[fusion71au]

2 hours ago, ellaosx said:

Thank you for clarification.

That said, , what does kernelXCPM now for?

 

AFAIK, Clover's original KernelXCPM setting works as before to enable Xnu kernel CPU Power Management for unsupported  SandyBridge, IvyBridge, Haswell, and Broadwell-E/EP processors - refer to @stevezheng's post and @Pike R. Alpha's blog for detailed background.

 

Check with @Slice and @Jief_Machak whether they intend to remove the above setting in future, given overlap with OC AppleXcpmExtraMsrs function.

 

 

Edited October 14, 2020 by fusion71au
correct links

[Matgen84]

2 minutes ago, fusion71au said:

 

AFAIK, Clover's original kernelXCPM setting works as before for for SandyBridge, IvyBridge, Haswell, and Broadwell-E/EP - refer to @stevezheng's post.

 

Check with @Slice and @Jief_Machak whether they intend to remove the above setting in future, given overlap with OC AppleXcpmExtraMsrs function.

 

 

 

@fusion71au the link to @stevezheng's post don't work: error 404.

[naiclub]

Spoiler

That's really what I think.
Left latest version master, commit 086e5682a.
Right version 5122-ocint-3009.3
Apparently There are 2 overlapping lines. It will load the BIOS file instead of the modified file.
The next question is, what will you do next?

[kushwavez]

My AGDP KextPatches aren't working anymore on Big Sur, but they are on Catalina. I used them for my VGA, HDMI outputs. I replaced them with agdpmod=vit9696 so it's okay, but I'm still trying to figure out why aren't they working on Big Sur. I used the name "com.apple.driver.AppleGraphicsDevicePolicy" on Catalina and the patch was successful. However on Big Sur it says:

10:395  0:006  OCAK: Failed to pk find com.apple.driver.AppleGraphicsDevicePolicy - Not Found
10:397  0:002  OC: Kernel patcher result 4 for com.apple.driver.AppleGraphicsDevicePolicy (com.apple.driver.AppleGraphicsDevicePolicy (AGDP Catalina 10.15.X Patch 5)) - Not Found

according to the debug.log. The Terminal shows that the name is correct: 

Executing: /usr/bin/kmutil showloaded
  156    0 0xffffff7f9e556000 0x9000     0x9000     com.apple.driver.AppleGraphicsDevicePolicy (6.1.24) F276DA99-0775-371F-BE61-C8E66EF429F9 <155 115 114 14 13 8 7 6 5 3 1>

Is it renamed or smth?

Patch:

Spoiler

<dict>
	<key>Comment</key>
	<string>AGDP Catalina 10.15.X Patch 1</string>
	<key>Disabled</key>
	<false/>
	<key>Find</key>
	<data>RURJRAA=</data>
	<key>InfoPlistPatch</key>
	<false/>
	<key>Name</key>
	<string>com.apple.driver.AppleGraphicsDevicePolicy</string>
	<key>Replace</key>
	<data>RURJSQA=</data>
</dict>

 

Here is the debug: debug.log

 

Anyway Clover V5124 is working well on my 3 computers (in my signature):

PC: Windows 10 OK, Catalina 10.15.7 OK, Big Sur Beta 9 OK (on separate HDD)

Spoiler

 

Notebook 1: Windows 10 OK, Catalina 10.15.7 OK, Big Sur Beta 9 OK, Fedora Linux OK (on same SSD)

Spoiler

 

Notebook 2: Windows 10 OK, Catalina 10.15.7 OK, Big Sur Beta 9 OK (on same SSD)

Spoiler

 

Catalina and Big Sur are running from the same EFI, nothing need to be modified.

Now waiting for the final Big Sur release!  

Edited October 14, 2020 by kushwavez

[Henry2010]

6 hours ago, Slice said:

I have to explain.

There are two main roads leading to purpose.

One, older way supporting by me and still working:

1. FakeSMC.kext+its plugins (IntelCpuMonitor.kext, RadeonMonitor.kext, ITEIT87x.kext, ACPIMonitor.kext, VoodooBatterySMC.kext, SMIMonitor.kext)

2. SMCHelper.efi driver

Second way, developed by vit9696 and Acidanthera group:

1. VirtualSMC.kext+its plugins (.....). It required Lilu.kext to be also installed.

2. VirtualSMC.efi driver.

 

Don't mix these different ways! Don't jump from one road to another on full speed!

 

Thanks, so that means FakeSMC* kexts will continue to work in 5123/5124+, even though OC includes VirtualSMC.efi. There's no conflict.

And if one decided to use VirtualSMC* kexts + Lilu, he wouldn't need to add an extra VirtualSMC.efi in the UEFI folder(?).

Edited October 14, 2020 by Henry2010