Jump to content
30960 posts in this topic

Recommended Posts

49 minutes ago, mifjpn said:

error: static_assert with no message is a C++17 extension [-Werror,-Wc++17-extensions]
  static_assert(sizeof(gSettings.ocBooterQuirks) == sizeof(mOpenCoreConfiguration.Booter.Quirks));

These are the small difference between Gcc and Xcode.

Committed.

 

@Matgen84 The second problem you had happens to me randomly. I'd say 1 time out of 10 on a x220 and 1 time out of 4 on VMWare. Don't know yet what it can be. Either a bug in OpenCore or a setting (slide setting ?).

I also have different behaviour after boot failed. Sometime even go back to Clover menu.

 

This beginning seems normal. I have the same, then a stop for 2 seconds.

50 minutes ago, mifjpn said:

BigSur20200917143013.jpg.392c2a6a5c57d6c0ca8b7a29edacbbd6.jpg

If you activate debug in Clover, you should also have an OpenCore debug log. It's in EFI folder, not in EFI/CLOVER folder.

 

Unfortunately, stopping here is something that never happens in my tests.

  • Thanks 1
3 minutes ago, Jief_Machak said:

These are the small difference between Gcc and Xcode.

Committed.

 

@Matgen84 The second problem you had happens to me randomly. I'd say 1 time out of 10 on a x220 and 1 time out of 4 on VMWare. Don't know yet what it can be. Either a bug in OpenCore or a setting (slide setting ?).

I also have different behaviour after boot failed. Sometime even go back to Clover menu.

 

This beginning seems normal. I have the same, then a stop for 2 seconds.

If you activate debug in Clover, you should also have an OpenCore debug log. It's in EFI folder, not in EFI/CLOVER folder.

 

Unfortunately, stopping here is something that never happens in my tests.

 

@Jief_Machak In my case, a stop more than 2 seconds :cry:  Thanks for your great job.

 

Opencore debug log attached.

 

opencore-2020-09-08-164352.txt

15 minutes ago, Jief_Machak said:

Try this one. It's not committed so take the attached EFI.

CloverX64.efi.zip

 

@Jief_Machak Same issue: long bootkernelextensions.kc stage, boot failed and black screen

 

 

debug.log

opencore-2020-09-08-164352.txt

Hi @Jief_Machak, just tested your last CLOVERX64.efi build, unfortunately I got same screen just like @mifjpn

tested on second system from my signature Acer Aspire A315-53

removed OCQUIRKS.EFI and used your KEXT from 11.zip

Edited by MICKHAEL
update
  • Sad 1

Having exact same stuck screen with my attempt to use "Oc Libs" with different project.
Got same result while loading Oc with bcfg cmd with no Bootstrap. Something like:
- Oc failed to get SmBios table from SystemConfigurationTable because it run to early, that I already resolved.
- They lost "Virtual FileHandle"?
- OpenRuntime is not running / running late after boot.efi already loaded?
* No idea whats going on there

  • Like 1
13 minutes ago, Jief_Machak said:

Sorry, wrong EFI sent. Try this one please.

CloverX64.efi.zip

 

Thanks @Jief_Machak Same issue: faster bootkernelextensions.kc stage (less than 15 seconds), boot failed and black screen.

Just a question: I press bar when Big Sur Installer is selected in Clover GUI, choose  "macOS with selected options", verbose mode start directly upon Gui. I don't understand why. See 
screenshot (Sorry for very bad English)

 

Spoiler

1.thumb.jpg.9f12caf8ab05f500d666affc7be23086.jpg

 

debug.log

opencore-2020-09-08-164352.txt

Always the same

having 3 EFI in the Z370 hack
EFI --> Clover bootloader default

EFI3 --> OpenCore

EFI2 --> bootloader windows

I inserted in EFI 2 the CLOVER folder of the EFI, and added the entries in the BIOS, so that I could be freer for experimentation.... 
but as written on top.. always doesn't get out of the initial memory manager do
 

  • Sad 1
1 hour ago, Matgen84 said:

Just a question: I press bar when Big Sur Installer is selected in Clover GUI, choose  "macOS with selected options", verbose mode start directly upon Gui. I don't understand why. See screenshot (Sorry for very bad English)

I saw that on your screenshots on I don't know why. A switch back to console we should do before starting a verbose boot, I guess.

 

6 minutes ago, doomking said:

clover can boot bigsur b6 on my laptop.

At least one can boot !!!

 

1 hour ago, mifjpn said:

Edit:I wait 5 min. It booted.

 But,login window moved to subdisplay(WhatEverGreen is not applied.)

 No sounds,
 USBPorts or USBmap is not applied,
 And SMBIOS is wrong.

Don't know why the long stop. OC says your kexts are injected fine.

 

The problem, from here, is that I have to compile an EFI, send it, someone try, send back log. That'll probably happens not only once ! Who's up for that ?

  • Like 3
4 minutes ago, Jief_Machak said:

Who's up for that ?

im ready)

 

maybe its better to send my working EFI for Catalina/ BigSur B2 and you check it/ adjust?

Edited by MICKHAEL
30 minutes ago, Jief_Machak said:

The problem, from here, is that I have to compile an EFI, send it, someone try, send back log. That'll probably happens not only once ! Who's up for that ?

 

@Jief_Machak I'm ready too. But tomorrow, I'm not at home.

Edited by Matgen84
21 hours ago, mifjpn said:

Edit:I wait 5 min. It booted.

Do you use a usb stick to try this "CloverOC". I wonder if the delay could just be log generated in a file on a slow usb stick.

Here is a version that properly integrate OC log into the Clover log file. If you put debug to false in config.plist, no log is saved and that might speed up the boot.

In this version, I've also uncommented a "SetDevices()" function that does some setup for device properties. See if that change anything with your missing devices.

 

Edited by Jief_Machak

By the way, did everyone who tried remember to copy your booter quirks in your config plist ?

 

Basically, in "KernelAndKextPatches" section, you should have something like :

		<key>OcFuzzyMatch</key>
		<false/>
		<key>OcKernelCache</key>
		<string>Auto</string>
		<key>OcQuirks</key>
		<dict>
			<key>AppleCpuPmCfgLock</key>
			<false/>
			<key>AppleXcpmCfgLock</key>
			<false/>
			<key>AppleXcpmExtraMsrs</key>
			<false/>
			<key>AppleXcpmForceBoost</key>
			<false/>
			<key>CustomSMBIOSGuid</key>
			<false/>
			<key>DisableIoMapper</key>
			<false/>
			<key>DisableLinkeditJettison</key>
 			<false/>
			<key>DisableRtcChecksum</key>
			<false/>
			<key>DummyPowerManagement</key>
			<false/>
			<key>ExternalDiskIcons</key>
			<false/>
			<key>IncreasePciBarSize</key>
			<false/>
			<key>LapicKernelPanic</key>
			<false/>
			<key>PanicNoKextDump</key>
			<false/>
			<key>PowerTimeoutKernelPanic</key>
			<false/>
			<key>ThirdPartyDrives</key>
			<false/>
			<key>XhciPortLimit</key>
			<false/>
		</dict>

If all quirks is false, you may omit the OcQuirks section.

 

Should I remove the Oc prefix I put in front ?

 

36 minutes ago, Jief_Machak said:

 

If all quirks is false, you may omit the OcQuirks section.

 

 

interesting. I omitted this OcQuirks step) I will check again. thanks

 

UPDATE
okay. not much changed,, I consulted also dortania for OCQUIRKS settings,
i attaching my working Clover EFI (Catalina/ Big Sur B2), please take a look and adjust it

Edited by MICKHAEL
update

In short :

- Quirks from section kernel from OC (containing "AppleCpuPmCfgLock"), must be in section KernelAndKextPatches with name OcQuirks

   OcFuzzyMatch and OcKernelCache beside this (not inside). They can be committed with this last EFI.

- Quirks from section booter from OC (containing "AvoidRuntimeDefrag"), must be under root, no section, with name OcQuirks

  MmioWhitelist, if defined, must be inside this Quirks section.

I attached my VM config.plist as an example.

 

Here is an EFI that contains panic if the Quirks sections are not found at the right place. Because you probably (like me) copy/paste the Quirks section for an OC config.plist, you may have swapped them. So I also put a panic if AppleCpuPmCfgLock quirks is not found, or if AvoidRuntimeDefrag is not found. That means you have to define them even if it's false. All the others are optional and default to false.

 

NOTE : the panics are temporary !!!

 

config.plist

 

 

Edited by Jief_Machak
  • Like 1

If you use 2 different CLOVER folder for your tests, you might like the new option.

In config.plist, section boot, key Debug, you can now use the value <string>scratch</string> to indicate that you want a log but you want it scratch at every boot.

 

It doesn't work if you use the same CLOVER folder (= if you use only one partition) and just ask your firmware to boot on different CloverX64.efi. That's because, of course, if the test crashes, the reboot on your working Clover will scratch the log.

 

I wouldn't recommend using only one CLOVER folder. It's dangerous because you will be using the same kext and driver folder, and most likely the same config.plist. If you modify something there and it's wrong, you may end up not being able to boot at all. We've seen a lot of people being locked out after a change like this.

I know Slice is using only one Clover folder, but he knows what's he doing ! Unless you know as much as him, be careful.

So, you have some choice with UEFI computer :

- if your firmware allows you to set the boot file create a new small FAT32 partition on you drive. Put you test Clover in there.

- use a USB stick. If you do that, put your working Clover on the USB stick and your test Clover on your drive. The reason is that USB stick are usually quite slow (using a new big USB3 stick would a waste, right ?). Booting a Clover that generates debug messages in a file could be quite sloooooooow. I mean reeeeeeeeally sloooooooooooooooooooow. If you have a working Clover folder, you can disable the log and the boot on a usb stick won't be much different.

- use a second drive. If you have more than one drive, use the EFI partition on one of the other drives.

 

With legacy computer, you can still use a USB stick or a second drive.

CloverX64.efi.zip

  • Like 1
3 minutes ago, iCanaro said:

this last Clover, doesn't even get to the GUI:
critical error, does not find ocquirks
whether it is present or absent in UEFI drivers, the result is always the same critical error

im not alone)

Yes, so correct your Quirks config in your config.plist

3 hours ago, Jief_Machak said:

In short :

- Quirks from section kernel from OC (containing "AppleCpuPmCfgLock"), must be in section KernelAndKextPatches with name OcQuirks

   OcFuzzyMatch and OcKernelCache beside this (not inside). They can be committed with this last EFI.

- Quirks from section booter from OC (containing "AvoidRuntimeDefrag"), must be under root, no section, with name OcQuirks

  MmioWhitelist, if defined, must be inside this Quirks section.

I attached my VM config.plist as an example.

 

×
×
  • Create New...