Jump to content

Clover problems report & features request


ErmaC
953 posts in this topic

Recommended Posts

12 hours ago, Slice said:

Test please new version with Pene corrections.

Clover_r5105.zip

thanks ! finally it showed SUCCESS for patches in the list : ) 

but the system rebooted in the end showing :

1914916906_ScreenShot2020-03-20at7_33_27PM.png.617041e9c3ba1532664323c0c5c65b0a.png

EDIT: 
i think it will also need Count, Limit, Skip fields of OC kernel patcher to make it work in clover too.
as few patches uses it.
OC_kernel_patches.plist 

Edited by XLNC
  • Like 2
Link to comment
Share on other sites

24 minutes ago, XLNC said:

Macos version is 10.15.3
kernel.zip

 

10 minutes ago, fabiosun said:

@XLNC also in Hs and Mojave same your results (and for my trx40 it is a correct behaviour for now)

 

 

very strange 

both 10.14.6 and 10.15.3 and 10.15.4 beta6 have same pattern for KxldUnmap patch. i checked it in kernel.

  • Like 1
Link to comment
Share on other sites

On 3/20/2020 at 10:35 PM, XLNC said:

Macos version is 10.15.3
kernel.zip

Was this problem solved ?

We need it for kernel patch via Clover bootloader in AMD hackintosh.

Currently only OpenCore can be used.

Link to comment
Share on other sites

14 hours ago, vector sigma said:

I guess of yes. There was a tiny mistake solved by Pene.

I have used Clover r.5106, but its Mask or ReplaceMask functions seemed not working for kernel patching yet.

Please test my config-test.plist in AMD FX hackintoshs.

config-test.plist

  • Like 1
Link to comment
Share on other sites

@Slice @vector sigma

 

Hack 4 signature, Clevo P170SM [HD4600 mobile]

 

With the clevo I entered a situation that currently has no way out. All releases of Clover following version 5098 do not load any ACPI tables and therefore the system immediately goes into lockout.

The problem would not be serious except that starting from the release of catalina 10.15.4beta6 Clover 5098 is unable to successfully complete the update.

All kexts and .efi drivers are updated to the latest versions. The config was completely recreated using the basic one included in the Clover iso 5107 using a plist editor

 

Video --> https://mega.nz/#!j3ICkS5b!c-rjnF0BQBtb492YSBkMZYVKDlCaAE1pfPjtvMgnc7M

 

Foto 1 --> https://mega.nz/#!f6BgQCKK!OVFT-XSIpoLAq4K7Dqq_shtE_M2K9OhLeUTx5xMo9gM

 

Foto 2 --> https://mega.nz/#!2z4hjJbY!3nlEHQCBfIVYjAI-mLQrYT4rgjKOS4lr0pZMqSSHo98

 

Foto 3 --> https://mega.nz/#!a2gRDZgZ!Rkls5DiXcl1Xq2QqBeOyvRMG-0UVTsPQJuvhCrjPtX8

1795074521_clevoacpi_IMG_20200325_170427.thumb.jpg.032961a9d4e4f304ccec892154a25e95.jpg

 

preboot.log_clover_5098.zip

preboot.log_clover_5107.zip

log_2.zip

LOG.zip

 

Edited by iCanaro
Link to comment
Share on other sites

 

new: with DSDT made to me by gengik84,

no VirtualSMC and all its plugins

but in its place FakeSMC included in the ISO of Clover 5107

I managed to finish the installation of the catalina update 10.15.4 19e266 also with Clover 5098

  • Like 1
Link to comment
Share on other sites

@iCanaro

With Clover 5098 you have to Kernel patches

2:337  0:000   - [00]: Disable panic kext logging on 10.14 Release kernel :: MatchOS: 10.14 :: data len: 6
2:337  0:000   - [01]: Disable panic kext logging on 10.13 Release kernel :: MatchOS: 10.13 :: data len: 6

while with Clover 5107 you have

2:908  0:000  KernelToPatch: 1 requested
2:908  0:000   - [00]: MSR 0xE2 _xcpm_idle instant reboot(c) Pike R. Alpha :: data len: 8

The difference is not related to Clover. This is the difference of two configs.

Link to comment
Share on other sites

@Slice True, those patches are in the used config, but have always been in the disabled mode. (are they to be enabled ???) As it is true that the situation of the various logs is not inherent to a static situation, but in evolution, so much so that in the end I managed to update to catalina 10.15.4 stable. If I write here, even with difficulty because of my lack of knowledge of the English language, it is not to make controversies or report false problems, but to bring to attention what is a real problem for me, namely that with all the versions following clover 5098 my clevo no longer starts. I am not a developer, so maybe I do not express myself in an appropriate way, it will not be the ACPI that is not loaded, but something that does not work in clover to start my system, exists, and I will be happy and happy when I can say that new release Clover works perfectly for me.

  • Like 1
Link to comment
Share on other sites

I downloaded CloverBootloader-master 84306ec last nite and compiled with 11.3 Xcode on MacOS using ./buildme option 6 for build all.

 

Everything compiled successfully, I got the CloverV2.zip, Clover.app_V1.16.pkg.zip and Clover_r.pkg.

 

I installed from Clover.app_v1.16.pkg and had it update Clover.  The logs showed that it did update Clover - and I could see the file date changes when I examined the EFI.

 

When I ran Hackintool, it showed a version of 0.

 

It booted up without problem, and with the Aero Theme, All looked normal Except there was no Clover version number in the lower right.

 

I downloaded the released version of the Clover.app_v1.16_r5107.pkg.zip - ran it, updated Clover, and ended up with 5107 showing as Clover's version.

 

1. What did I do wrong?

 

2. For the intermediate changes between the releases, would it be possible to display for the version something like 5107.84306ec (showing the GitHub commit along with the version)?

 

3. Is there a script that would download everything from Github like:
      git clone https://github.com/CloverHackyColor/CloverBootloader
where it would be possible to have the option 9 (update clover) pull down any further commits from GitHub (while retaining the tool builds in the toolchain)?

 

When I try option 9, it says something about not being a valid git repository, something like that - so I know I haven't set it up correctly as a git repository.

Link to comment
Share on other sites

9 hours ago, exquirentibus said:

1. What did I do wrong?

 

2. For the intermediate changes between the releases, would it be possible to display for the version something like 5107.84306ec (showing the GitHub commit along with the version)?

 

3. Is there a script that would download everything from Github like:
      git clone https://github.com/CloverHackyColor/CloverBootloader
where it would be possible to have the option 9 (update clover) pull down any further commits from GitHub (while retaining the tool builds in the toolchain)?

 

When I try option 9, it says something about not being a valid git repository, something like that - so I know I haven't set it up correctly as a git repository.

 

  1. everythings. 
  2. already exist but you download a zip directory instead of the repository.
  3. you write the correct command and you did not even use it?:| (git clone https://github.com/CloverHackyColor/CloverBootloader). After that just double click click on buildme.

 

Link to comment
Share on other sites

11 hours ago, iCanaro said:

@Slice True, those patches are in the used config, but have always been in the disabled mode. (are they to be enabled ???) As it is true that the situation of the various logs is not inherent to a static situation, but in evolution, so much so that in the end I managed to update to catalina 10.15.4 stable. If I write here, even with difficulty because of my lack of knowledge of the English language, it is not to make controversies or report false problems, but to bring to attention what is a real problem for me, namely that with all the versions following clover 5098 my clevo no longer starts. I am not a developer, so maybe I do not express myself in an appropriate way, it will not be the ACPI that is not loaded, but something that does not work in clover to start my system, exists, and I will be happy and happy when I can say that new release Clover works perfectly for me.

I can only say that you mixed a lot of stuff and conditions where too different. You take Slice's words in the wrong way.. not sure why :) but the log and your screenshot show warnings about ACPI tables... so the tables were loaded :).

Edited by vector sigma
Link to comment
Share on other sites

1 hour ago, vector sigma said:

 

  1. everythings. 
  2. already exist but you download a zip directory instead of the repository.
  3. you write the correct command and you did not even use it?:| (git clone https://github.com/CloverHackyColor/CloverBootloader). After that just double click click on buildme.

 

I did try that script yesterday.  It pulled in the latest from github, but ... something went wrong.  I ended up having to download the zipped 5107 source to get it to compile, and then I couldn't update using git.

 

One tiny mistaken assumption, that a more experience Apple developer would never make and is so obvious, can be frustrating to someone trying to learn coding on MacOS!!

 

I'll figure it out.  I realize that people struggling with the very basics can be Very annoying.

 

Compiling the toolchain tools seems to take the most time. 

 

1. Can I assume that the entire toolchain folder stays the same and can be saved and restored while testing - from build to build? 

 

2. Is the missing clover version normal when building a non-final commit?

 

Thanks!

Link to comment
Share on other sites

2 hours ago, exquirentibus said:

 

1. Can I assume that the entire toolchain folder stays the same and can be saved and restored while testing - from build to build? 

 

2. Is the missing clover version normal when building a non-final commit?

  1. you need to build it once. Pay attention to not move the repository (i.e. on a different path) because command lines like gettext hard coded and expect paths to be where it knows.
  2. Not sure what do you mean, but the repository include tags which gives the last revision. But since git use the SHA (or hash), a revision can include all the commits made until developers decide that a particular commit is promoted as "stable" ..to be released.. with a new revision...
  3. The repository has no brain... no need to say that your clone needs to be updated to have last code.
  4. As per 3., developers add code, works with others from around the world ... and often the last code can potentially be an unfinished job (a work in progress)..

 

Edited by vector sigma
gettex + t
Link to comment
Share on other sites

17 minutes ago, vector sigma said:
  1. you need to build it once. Pay attention to not move the repository (i.e. on a different path) because command lines like gettext hard coded and expect paths to be where it knows.
  2. Not sure what do you mean, but the repository include tags which gives the last revision. But since git use the SHA (or hash), a revision can include all the commits made until developers decide that a particular commit is promoted as "stable" ..to be released.. with a new revision...
  3. The repository has no brain... no need to say that your clone needs to be updated to have last code.
  4. As per 3., developers add code, works with others from around the world ... and often the last code can potentially be an unfinished job (a work in progress)..

 

Thanks for taking the time to reply.  Definitely a pain to deal with someone like me who is a bit clueless, when there's so much stress with the virus.

 

For #2. what I'm referring to is that when I see the clover screen in Aero, it shows the Clover version # in the lower right.   Hackintool also shows the version installed, as well as I believe the Clover.app.

 

This version, after compiling the latest commits last nite showed zero in those programs, and showed just an empty space at the bottom of the Aero theme in Clover.  

 

I'm just curious if this was normal for non-release builds?   I do think it would be nice to have the major version (like 5107) and the minor version (like the hex code showing on gitHub commit) so that it's easy to identify what's installed (I have several SSD's for testing, it would be great if it showed the commit version of the build as well as the last main release version - so it's easy to tell them apart).

Link to comment
Share on other sites

11 hours ago, vector sigma said:

I can only say that you mixed a lot of stuff and conditions where too different. You take Slice's words in the wrong way.. not sure why :) but the log and your screenshot show warnings about ACPI tables... so the tables were loaded :).

 

Slice is absolutely perfect in his statements, what he says is true; but the question remains for me, until Clover 5098 everything OK, from all subsequent releases of Clover, it no longer starts, it immediately stops as per screenshots. So if someone simply has even an idea of what to set in the config or something else to avoid this good.

Link to comment
Share on other sites

1 hour ago, iCanaro said:

So if someone simply has even an idea of what to set in the config or something else to avoid this good.

Sorry but it seemed in your next post you were saying that you had solved the problem using fakesmc instead of virtualsmc. Thinking..

 

What happens if you don't drop the cpupm table?

Edited by vector sigma
Link to comment
Share on other sites

×
×
  • Create New...