Clover General discussion

[apianti]

There you go. I tried with multipile DSDT, plist etc.. but the same freeze. The only thing I can say is that the phone "mounts" or represent a virtual CDROM for Mac even in charging only mode on which you can find Android File transfer... I think that the virtual CDROM that phone has on USB connection is the problem with clover

 

Here you can see how it is recognized by macOS:

 

Yeah, it definitely appears to be the USB volume on the phone:

1:351  0:003          USB volume
1:352  0:001  - [03]: Volume: PciRoot(0x0)\Pci(0x14,0x0)\USB(0x8,0x1)
0:100  0:100  MemLog inited, TSC freq: 3691449556

I am wondering if the APFS driver is causing issues by writing to the console.... 

[Slice]

 

@slice @PMHeart

 

I suggest to optimize Chinese descriptions in Clover Install GUI such as Driver64UEFI.

If descriptions have some mistake please fix it.

Thanks.

 

AptioInputFix

Vit9696编写的针对FileVault2启动界面的键盘和鼠标输入设备的支持

 

 

AptioMemoryFix

Vit9696自OsxAptioFix2驱动优化而来的新驱动，支持更多新特性，例如：

 

自动为boot.efi寻找最适合的内存地址，避免启动错误

当slide值不能被使用的时候提供对KASLR的支持

新增系统在低内存地址下的安全模式的支持

确保系统不会出现slide溢出的问题

尝试修复更多的内存分布问题

支持硬件NVRAM

优化一些休眠的问题（暂时不稳定）

（请勿与其余内存修复驱动同时使用）

 

 

CsmVideoDxe-64

提供对CSM模块的支持，某些主板若要安装非UEFI系统或者非安全启动的系统，例如Windows7，Linux等设备，需要开启CSM模块，此时需要加入该驱动以修复CloverGUI的显示问题

Clover推荐关闭CSM模块启用原生GOP显示模块

 

 

EmuVariableUefi-64

macOS使用NVRAM存储一些系统变量，大部分的UEFI主板在配合合适的Aptio驱动后支持原生硬件NVRAM，但是少部分主板不支持NVRAM或者NVRAM的支持有问题，此时建议加入该驱动，该驱动通过开机时加载位于EFI分区内的nvram.plist内容到nvram中，以模拟NVRAM支持

需要注意的是，是用此驱动，需要勾选“安装RC scripts到目标磁区”选项才有效

 

 

Fat-64

可选择的64位FAT文件系统的支持

 

 

OsxAptioFixDrv-64

Dmazar编写的针对UEFI固件的内存问题修复的驱动，对休眠支持不完善

（请勿与其余内存修复驱动同时使用）

 

 

OsxAptioFix2Drv-64

Dmazar编写的针对UEFI固件的内存问题修复的驱动，在1代基础之上完善了休眠等高级功能的支持，部分机型需要手动设置slide值

（请勿与其余内存修复驱动同时使用）

 

 

OsxAptioFix3Drv-64

Vit9696等作者在OsxAptioFix2Drv-64的基础之上进行了优化，修复了大多数新设备的NVRAM支持，该驱动在部分机型依然需要手动设置slide值

（请勿与其余内存修复驱动同时使用）

 

 

OsxFatBinaryDrv-64

可选择的64位FAT文件系统的支持

 

 

OsxLowMemFixDrv-64

针对UEFI固件内存问题修复的驱动简化版

（请勿与其余内存修复驱动同时使用）

 

 

PartitionDxe-64

支持非常态的分区配置，如苹果混合分区或Apple分区图。

 

 

UsbkbDxe-64

FileVault2启动界面的键盘输入设备的支持

 

 

UsbMouseDxe-64

FileVault2启动界面的鼠标输入设备的支持

 

Please, edit these files with TextEdit and upload the result

Архив.zip

[Guest]

I am sorry in advance with this little offtopic

Is there someone using this patch in 10.13.3 (okrasit patches for EIST useful to boost all cores if EIST is enabled in bios)

 

f

89d8c1e0 084863d0 

r

b800ff00 004863d0 

comment

xcpm_assert_perf_max © okrasit

 

it seems is not working in 10.13.4 beta x

 

from this link  kernel.txt

 

https://we.tl/DCyFxHdsoF 

[gujiangjiang]

Please, edit these files with TextEdit and upload the result

Архив.zip

 

I found po folder and Localizable.strings are also lack many description and i dont know how to add it to Localizable.strings with these new adding efi drivers.

[Funky frank]

I have a question: Windows can and does load cpu microcode to the cpu on boot, so not permanent flashing.

 

Does macOS do the same, and if so, what kext or command is responsible for that? Or will be the cpu microcode inside the mac efi firmware? 

[apianti]

I have a question: Windows can and does load cpu microcode to the cpu on boot, so not permanent flashing.

 

Does macOS do the same, and if so, what kext or command is responsible for that? Or will be the cpu microcode inside the mac efi firmware? 

 

Windows selectively does this, depending on for what, for this they told OEMs that they need to provide firmware updates for the microcode updates. Apple uses firmware updates as well. I'm guessing that if your firmware has had a microcode update before you could probably get the module from the firmware and figure out how it works.

 

There you go. I tried with multipile DSDT, plist etc.. but the same freeze. The only thing I can say is that the phone "mounts" or represent a virtual CDROM for Mac even in charging only mode on which you can find Android File transfer... I think that the virtual CDROM that phone has on USB connection is the problem with clover

 

Here you can see how it is recognized by macOS:

 

Can you please try this clover and see what happens?

 

EDIT: Removed binary.

[maxb2000]

Windows selectively does this, depending on for what, for this they told OEMs that they need to provide firmware updates for the microcode updates. Apple uses firmware updates as well. I'm guessing that if your firmware has had a microcode update before you could probably get the module from the firmware and figure out how it works.

 

 

Can you please try this clover and see what happens?

 

Same stuff: Freeze on APFS log

[apianti]

Same stuff: Freeze on APFS log

 

Did you replace \EFI\BOOT\BOOTX64.efi too? It should not have been able to print anything out.

[maxb2000]

Did you replace \EFI\BOOT\BOOTX64.efi too? It should not have been able to print anything out.

 

You only attached CLOVERX64.efi to the post. 

[bronxteck]

they should be the same binary they just have different names. so rename it in the boot folder

[maxb2000]

Did you replace \EFI\BOOT\BOOTX64.efi too? It should not have been able to print anything out.

 

they should be the same binary they just have different names. so rename it in the boot folder

 

Done, I downloaded bootx64.efi and copied it to the Clover folder and rename it ! Still same freeze and the APFS log is still there.

 

Btw: why we have 2 Clover binaries ? instead of just: \EFI\Boot\BOOTX64.efi ?

[apianti]

Done, I downloaded bootx64.efi and copied it to the Clover folder and rename it ! Still same freeze and the APFS log is still there.

 

Btw: why we have 2 Clover binaries ? instead of just: \EFI\Boot\BOOTX64.efi ?

 

Because clover is at \EFI\CLOVER\CLOVERX64.efi and \EFI\BOOT\BOOTX64.efi is the path for the default boot loader of volume, and could be anything you want it to be, like windows or linux.

 

So the APFS driver must actually locate the EFI_SIMPLE_TEXT_OUTPUT_PROTOCOL to write to output instead of using the ConOut from SystemTable.... Oh jeez.

 

EDIT: Whenever it writes to the console that causes the graphics to switch back to 1024x768. But also the graphics are doing stuff, and especially when using CsmVideo and it needs to reconnect. Does ConSplitter need to be used immediately if APFS driver is being loaded?

[maxb2000]

Graphics switch to 1024 x 768 when entering UEFI Shell also. I have Windows 8 WHQL set in UEFI settings so it loads the UEFI GOP Driver. This way, the BGRT (The logo of the OEM) is at native resolution, in my case 1920 x 1080.

 

The thing is that I don't understand why UEFI Shell and APFS driver switch it to 1024 x 768. 

[arsradu]

Graphics switch to 1024 x 768 when entering UEFI Shell also. I have Windows 8 WHQL set in UEFI settings so it loads the UEFI GOP Driver. This way, the BGRT (The logo of the OEM) is at native resolution, in my case 1920 x 1080.

 

The thing is that I don't understand why UEFI Shell and APFS driver switch it to 1024 x 768. 

 

CSM OFF in BIOS? Just double checking.

[Trung_Nguyen]

Unfortunately, no, those keys are being consumed by the input buffer for FileVault2 for startup keys and stuff.... For now you will have to put at least Timeout=1, in the future I guess that protocol will need to be located and checked for this purpose.... I've no idea who's going to do that....

 

EDIT: I should also point out the startup combinations for mac don't work because that is part of the mac firmware and is not implemented in clover.... Yet, I guess.

I have worked for that and successful! Now, how could I commit&Push it to sourceforge?

[apianti]

Graphics switch to 1024 x 768 when entering UEFI Shell also. I have Windows 8 WHQL set in UEFI settings so it loads the UEFI GOP Driver. This way, the BGRT (The logo of the OEM) is at native resolution, in my case 1920 x 1080.

 

The thing is that I don't understand why UEFI Shell and APFS driver switch it to 1024 x 768. 

 

They switch because the console has a specific resolution for the size of character and by columns and rows. So anytime the console is written to it switches to the mode it needs to write out the characters to the graphics. The shell is fine because it is an application and the only thing using the console/graphics during that time (everything should be reset when you exit back into clover). APFS is a driver, it acts asynchronously, so as stuff happens with the volumes handled by the driver, it prints out messages, which if you are drawing on the screen or otherwise using something to do with graphics, this switch is a problem because it probably preempts the current task of drawing, making the graphics buffer no longer valid but it's still drawing inside that buffer. As far as CsmVideo, I noticed this because someone opened a ticket for this exact problem but it was spamming in the debug.log that it was blocking the switch to 1024x768, or it just crashed at trying to reconnect.

 

won't understand your claims, adjust.

 

Don't know what this means, I understood him perfectly fine, this though, not as sure....

 

graphics isn't created

 

If graphics weren't created then there would not be anything on screen, not even APFS log. This is related to APFS driver because it directly happened after moving over to APFS after not having used it...

 

I have worked for that and successful! Now, how could I commit&Push it to sourceforge?

 

Post the diff patch here, for review.

[maxb2000]

CSM OFF in BIOS? Just double checking.

 

Yes, I always turn off CSM and sometimes I enable fast boot. The second question is. Can we turn off the APFS log in Clover ? 

[apianti]

Yes, I always turn off CSM and sometimes I enable fast boot. The second question is. Can we turn off the APFS log in Clover ? 

 

Never enable fast boot. It means your firmware will never be reinitialized fully when you restart, and has no effect besides not loading some stuff from cold boot. Since you don't know what's it's reinitializing, it could leave a bunch of stuff present that you don't want and can cause issues, especially with ACPI. I tried to turn off the APFS log, that's what that binary I posted for you to try did. The only other thing is completely disabling console output, or splitting the console from the graphics output, either way you are left without console output unless clover makes its own console that doesn't need to switch..... Hmmmmm..... Hmmmmmmmmmmmmmmmmmm........................ Hmmmmmmmmmmmmmmmmmmmmmmmmmmmmmm......................................

[arsradu]

Yes, I always turn off CSM and sometimes I enable fast boot. The second question is. Can we turn off the APFS log in Clover ? 

 

One way here: http://www.insanelymac.com/forum/topic/327584-apfsefi-without-verbose-boot/page-5?do=findComment&comment=2532527

Another way here: http://www.insanelymac.com/forum/topic/291655-ozmosis/?p=2498575 (requires Patcho).

 

I would leave Fast Boot off. I had issues with it, since it was skipping stuff I didn't want skipped, such as the Clover UI.

 

I suppose you don't have CSMVideo driver in Clover, right? You don't need it anyway.  And it could create issues if it's installed. Again, just trying to troubleshoot the issue here. Since it looks like you did what you had to do. And still...it doesn't seem to work for you, for some reason.

[apianti]

One way here: http://www.insanelymac.com/forum/topic/327584-apfsefi-without-verbose-boot/page-5?do=findComment&comment=2532527

Another way here: http://www.insanelymac.com/forum/topic/291655-ozmosis/?p=2498575(requires Patcho).

 

I would leave Fast Boot off. I had issues with it, since it was skipping stuff I didn't want skipped, such as the Clover UI.

 

I suppose you don't have CSMVideo driver in Clover, right? You don't need it anyway.  And it could create issues if it's installed. Again, just trying to troubleshoot the issue here. Since it looks like you did what you had to do. And still...it doesn't seem to work for you, for some reason.

 

http://www.insanelymac.com/forum/topic/327584-apfsefi-without-verbose-boot/page-5?do=findComment&comment=2582519

 

The patched driver is also in there, I forgot about that, duh.

[arsradu]

http://www.insanelymac.com/forum/topic/327584-apfsefi-without-verbose-boot/page-5?do=findComment&comment=2582519

 

The patched driver is also in there, I forgot about that, duh.

 

That too.

[maxb2000]

Never enable fast boot. It means your firmware will never be reinitialized fully when you restart, and has no effect besides not loading some stuff from cold boot. Since you don't know what's it's reinitializing, it could leave a bunch of stuff present that you don't want and can cause issues, especially with ACPI. I tried to turn off the APFS log, that's what that binary I posted for you to try did. The only other thing is completely disabling console output, or splitting the console from the graphics output, either way you are left without console output unless clover makes its own console that doesn't need to switch..... Hmmmmm..... Hmmmmmmmmmmmmmmmmmm........................ Hmmmmmmmmmmmmmmmmmmmmmmmmmmmmmm......................................

 

 

Good to know that. One more question. If I enable timeout 0 in clover and I press any key why doesn't stop the boot like it should ?

[Gigamaxx]

http://www.insanelymac.com/forum/topic/327584-apfsefi-without-verbose-boot/page-5?do=findComment&comment=2582519

 

The patched driver is also in there, I forgot about that, duh.

 

@apianti, is there a way to disable IntelCPUMonitor in clover during the boot process?    We have an issue with the latest High Sierra 13.4 betas which never arose before in High Sierra for AMD CPUs using our custom kernels.    It only seems to affect HS 13.4 and may be related to Apples fixes for Spectre and meltdown bug fixes?   Any thoughts or suggestions on how to find this and disable it for AMD CPU users?  

 

My initial Google searches point towards FakeSMCkext and Ivy Bridge from 2012.    Is there any terminal command that would be able to locate IntelCPUMonitor in HS 13.4 so I could try removing it from the software for testing?      Any help would be greatly appreciated.

 

[arsradu]

@apianti, is there a way to disable IntelCPUMonitor in clover during the boot process?    We have an issue with the latest High Sierra 13.4 betas which never arose before in High Sierra for AMD CPUs using our custom kernels.    It only seems to affect HS 13.4 and may be related to Apples fixes for Spectre and meltdown bug fixes?

No idea on the AMD CPUs....but I doubt the issues come from the Spectre fix.

 

For as far as I know, the fix has been applied in 10.13.2 Supplemental Update. So...unless there's another iteration on that, I'm not sure they're related.

[Gigamaxx]

No idea on the AMD CPUs....but I doubt the issues come from the Spectre fix.

 

For as far as I know, the fix has been applied in 10.13.2 Supplemental Update. So...unless there's another iteration on that, I'm not sure they're related.

 

The 13.4 betas have a new separate Immutable kernel in Prelinkedkernels folder that links to the main partition in Scripts, and Payloads folders which are new additions to HS software.    These bugs are kernel related and the fixes are focused on preventing kernel (kernel extensions) penetration by outside users.  

 

Since these have been added in 13.4 the IntelCPUMonitor errors have occurred it may be unrelated but if Apple is trying to apply fixes for Intel CPUs they may be detecting the CPU type in order to add the proper fixes.   Fixes for Skylake are probably different from haswell etc...    

 

Does anyone have any idea as to where IntelCPUMonitor would be located in OS?