OpenCore General Discussion

[topolino1]

7 hours ago, bbritton said:

After updating to 0.5.8 my Catalina Dynamic desktop has lines and spots on it.  Seems the graphics are screwed up a bit.  It is only on the Catalina Dynamic desktop though as all other desktops apps and web pages are good.  I used to have this issue with clover and was very happy when it disappeared with open core 0.5.7. 

 

If you use Intel 630 graphics. I think you have PciRoot (0x0) / Pci (0x2,0x0) set incorrectly.
This is probably a problem.

[bbritton]

1 hour ago, topolino1 said:

 

If you use Intel 630 graphics. I think you have PciRoot (0x0) / Pci (0x2,0x0) set incorrectly.
This is probably a problem.

Thanks but I do not use the Intel 630 to drive my screen, I use a Vega 64.  I only. intend to use the Intel 630 graphics for H265 etc.

[MacNB]

1 hour ago, markl18 said:

hi does any body knows where are the resources for 0.5.9 yes it is out for couple of weeks now it is very different they apparently took a very different approach to devices there is a shell script that quires your devices and generates devices.plist that being used by the process a much smarter approach than everyone doing who knows what with their devices

6 DeviceProperties 6.1 Introduction

Device configuration is provided to macOS with a dedicated buffer, called EfiDevicePathPropertyDatabase. This buffer is a serialised map of DevicePaths to a map of property names and their values.

Property data can be debugged with gfxutil. To obtain current property data use the following command in macOS:

ioreg -lw0 -p IODeviceTree -n efi -r -x | grep device-properties | sed 's/.*<//;s/>.*//' > /tmp/device-properties.hex &&
gfxutil /tmp/device-properties.hex /tmp/device-properties.plist && cat /tmp/device-properties.plist

6.2 Properties

1.

2.

6.3

Add
Type: plist dict
Description: Sets device properties from a map (plist dict) of deivce paths to a map (plist dict) of variable names and their values in plist metadata format. Device paths must be provided in canonic string format (e.g. PciRoot(0x0)/Pci(0x1,0x0)/Pci(0x0,0x0)). Properties will only be set if not present and not blocked.

Note: Currently properties may only be (formerly) added by the original driver, so unless a separate driver was installed, there is no reason to block the variables.

Block
Type: plist dict
Description: Removes device properties from a map (plist dict) of deivce paths to an array (plist array) of variable names in plist string format.

 

 Your first question/statement does not make sense.

 

If you are asking where the Resources folder is, then it is in the same place as it always was here.

 

But that is totally unrelated to DeviceProperties.

Where did you read about "very different approach" and device.plist ????

It is not mentioned in the Configuration PDF you have quoted.

Nothing has changed as far as I can tell with 0.5.9

 

Are you asking a question about Device Properties or making a suggestion ?

[bbritton]

9 hours ago, bbritton said:

After updating to 0.5.8 my Catalina Dynamic desktop has lines and spots on it.  Seems the graphics are screwed up a bit.  It is only on the Catalina Dynamic desktop though as all other desktops apps and web pages are good.  I used to have this issue with clover and was very happy when it disappeared with open core 0.5.7. 

 

Any thoughts of what could be the issue?  I have attached my config.plist and screen shot.

 

Thanks 

Config.plist

Just figured out the problem.  Not sure which Item fixed the graphics issue but I tried the "sanity Checker" from the OpenCore guide and it advised me to make three changes:

Booter
MmioWhitelist
Quirks

ProtectUefiServices = No.  I had set to Yes

SyncRuntimePermissions = Yes     I had set to No

NVRAM

WriteFlash = Yes     I had set to No

 

The Dynamic desktop works now.

 

[topolino1]

57 minutes ago, bbritton said:

Thanks but I do not use the Intel 630 to drive my screen, I use a Vega 64.  I only. intend to use the Intel 630 graphics for H265 etc.

 

If you use Vega, the more you have a bad section PciRoot (0x0) / Pci (0x2,0x0).

 

Here you have described exactly how it should be.

https://dortania.github.io/OpenCore-Desktop-Guide/config.plist/coffee-lake.html#deviceproperties

 

You generally have some errors in the config.

 

[markl18]

1 hour ago, MacNB said:

 

 Your first question/statement does not make sense.

 

If you are asking where the Resources folder is, then it is in the same place as it always was here.

 

But that is totally unrelated to DeviceProperties.

Where did you read about "very different approach" and device.plist ????

It is not mentioned in the Configuration PDF you have quoted.

Nothing has changed as far as I can tell with 0.5.9

 

Are you asking a question about Device Properties or making a suggestion ?

I thought that that what happens they look into that plist file and add it to the internal map 

just juding by these words

Note: Currently properties may only be (formerly) added by the original driver, so unless a separate driver was installed, there is no reason to block the variables.

Block
Type: plist dict
Description: Removes device properties from a map (plist dict) of deivce paths to an array (plist array) of variable names in plist string format.

 

that makes it sounds like they pick that plist file and add it to the internal map of stuff they even provide shell script to create the file atomaticly if not then they should and yes the resources file came empty thanks for the location

 

Edited May 6, 2020 by markl18

[bbritton]

35 minutes ago, topolino1 said:

 

If you use Vega, the more you have a bad section PciRoot (0x0) / Pci (0x2,0x0).

 

Here you have described exactly how it should be.

https://dortania.github.io/OpenCore-Desktop-Guide/config.plist/coffee-lake.html#deviceproperties

 

You generally have some errors in the config.

 

Thanks, I edited the Config and attached.  i believe it is correct now.  I believe I copied and pasted too many times before and got some extra lines in code.

But now on my login page I get some vertical green lines as per the pic attached.  But the desktop is ok now.

 

I have attached the new config file as well.

 

Config.plist

[MacNB]

53 minutes ago, markl18 said:

 

 

You have completely misunderstood that whole section 6.1.

That EXAMPLE mentioned in the Section 6.1 regarding the use of the ioreg command from terminal is for you to be able to retrieve the device properties from the device tree of your system. That example command saves the list to a plist file as CONVENIENT way to see what you have (for debugging purposes).

OC does nothing with that plist file.

 

As Obi-Wan Kenobi once said "These aren't the droids you're looking for...move along"

[topolino1]

The guide has information on how the section should look correctly.

Start with this!

 

You have nothing correctly.

The section you entered is responsible for Intel 630 graphics:

<key>PciRoot(0x0)/Pci(0x2,0x0)</key>
            <dict>
                <key>AAPL,ig-platform-id</key>
                <data>BwCbPg==</data>
                <key>framebuffer-con0-alldata</key>
                <data>AQAJAAAEAADHAwAA</data>
                <key>framebuffer-con0-enable</key>
                <data>AQAAAA==</data>
                <key>framebuffer-con1-alldata</key>
                <data>AgAKAAAEAADHAwAA</data>
                <key>framebuffer-con1-enable</key>
                <data>AQAAAA==</data>
                <key>framebuffer-con2-alldata</key>
                <data>AwQIAAAIAADHAwAA</data>
                <key>framebuffer-con2-enable</key>
                <data>AQAAAA==</data>
                <key>framebuffer-patch-enable</key>
                <data>AQAAAA==</data>
                <key>framebuffer-stolenmem</key>
                <data>AAAwAQ==</data>
            </dict>
        </dict>

 

 

5 hours ago, bbritton said:

Thanks, I edited the Config and attached.  i believe it is correct now.  I believe I copied and pasted too many times before and got some extra lines in code.

But now on my login page I get some vertical green lines as per the pic attached.  But the desktop is ok now.

 

I have attached the new config file as well.

 

Config.plist

 

[markl18]

4 hours ago, MacNB said:

 

You have completely misunderstood that whole section 6.1.

That EXAMPLE mentioned in the Section 6.1 regarding the use of the ioreg command from terminal is for you to be able to retrieve the device properties from the device tree of your system. That example command saves the list to a plist file as CONVENIENT way to see what you have (for debugging purposes).

OC does nothing with that plist file.

 

As Obi-Wan Kenobi once said "These aren't the droids you're looking for...move along"

then at least this should be their option of 'proper tree' options to populate the plist file like it does

[bbritton]

5 hours ago, topolino1 said:

The guide has information on how the section should look correctly.

Start with this!

 

You have nothing correctly.

The section you entered is responsible for Intel 630 graphics:

<key>PciRoot(0x0)/Pci(0x2,0x0)</key>
            <dict>
                <key>AAPL,ig-platform-id</key>
                <data>BwCbPg==</data>
                <key>framebuffer-con0-alldata</key>
                <data>AQAJAAAEAADHAwAA</data>
                <key>framebuffer-con0-enable</key>
                <data>AQAAAA==</data>
                <key>framebuffer-con1-alldata</key>
                <data>AgAKAAAEAADHAwAA</data>
                <key>framebuffer-con1-enable</key>
                <data>AQAAAA==</data>
                <key>framebuffer-con2-alldata</key>
                <data>AwQIAAAIAADHAwAA</data>
                <key>framebuffer-con2-enable</key>
                <data>AQAAAA==</data>
                <key>framebuffer-patch-enable</key>
                <data>AQAAAA==</data>
                <key>framebuffer-stolenmem</key>
                <data>AAAwAQ==</data>
            </dict>
        </dict>

 

 

 

I have put your code into my config.  It seems to be the same as I have had at one time before.  The dynamic desktop again now has the lines like originally my problem.  I have attached the latest config.

Config.plist

[topolino1]

@bbritton You still have mistakes.

 

Here you have described exactly how it should be.

https://dortania.github.io/OpenCore-Desktop-Guide/config.plist/coffee-lake.html#deviceproperties

[vit9696]

On 5/4/2020 at 7:38 PM, Sander Spilleman said:

Hi,

 

I've just upgraded to OC 0.5.8. I'm on macOS 10.15.4 on a Asus Prime Z390-A. Over time I've collected some stuff in my config related to RTC patching, which is needed as otherwise I have that F1 POST / Safe mode {censored}. I might have some stuff in my config that may not be needed anymore. I'd like to get some support to get confirmed what is still needed and want I can remove by now.

 

My config contains the following

Kernel --> Quirks --> DisableRtcChecksum set to TRUE (added in update to OC 0.5.8)

UEFI --> ProtocolOverrides --> AppleRtcRam set to TRUE  (added in update to OC 0.5.8)

NVRAM --> Add --> 7C436110-AB2A-4BBB-A880-FE41995C9F82 --> wake-failure set to 5 bytes 0 (added after 10.15.4: https://github.com/acidanthera/bugtracker/issues/765)

ACPI --> Add --> SSDT-AWAC.aml (per vanilla desktop guide)

 

ACPI --> Patch as follows: (I believe this was generated by SSDTTime)

 

And Kernel --> Add --> Patch the following 2: (found these in here... somewhere)

 

Is there something I can get rid of??

 

Thank you so much!

 

Sander

1. ProtocolOverrides — you do not need to override AppleRtcRam as you do not have Apple hardware. Leave it as false.

2. wake-failure can and should be removed.

3. AppleRTC kernel patches can be removed (they are replaced by a quirk).

4. You forgot to mention rtc-blacklist, but I should remind you that it needs to be set to 58 59 for Z390-like.

5. I don't know whether you need an ACPI patch.

On 5/4/2020 at 7:38 PM, Sander Spilleman said:

Hi,

 

I've just upgraded to OC 0.5.8. I'm on macOS 10.15.4 on a Asus Prime Z390-A. Over time I've collected some stuff in my config related to RTC patching, which is needed as otherwise I have that F1 POST / Safe mode {censored}. I might have some stuff in my config that may not be needed anymore. I'd like to get some support to get confirmed what is still needed and want I can remove by now.

 

My config contains the following

Kernel --> Quirks --> DisableRtcChecksum set to TRUE (added in update to OC 0.5.8)

UEFI --> ProtocolOverrides --> AppleRtcRam set to TRUE  (added in update to OC 0.5.8)

NVRAM --> Add --> 7C436110-AB2A-4BBB-A880-FE41995C9F82 --> wake-failure set to 5 bytes 0 (added after 10.15.4: https://github.com/acidanthera/bugtracker/issues/765)

ACPI --> Add --> SSDT-AWAC.aml (per vanilla desktop guide)

 

ACPI --> Patch as follows: (I believe this was generated by SSDTTime)

 

And Kernel --> Add --> Patch the following 2: (found these in here... somewhere)

 

Is there something I can get rid of??

 

Thank you so much!

 

Sander

1. ProtocolOverrides — you do not need to override AppleRtcRam as you do not have Apple hardware. Leave it as false.

2. wake-failure can and should be removed.

3. AppleRTC kernel patches can be removed (they are replaced by a quirk).

4. You forgot to mention rtc-blacklist, but I should remind you that it needs to be set to 58 59 for Z390-like.

5. I don't know whether you need an ACPI patch.

 

As for ACPI patches and

Edited May 7, 2020 by vit9696

[LSP414]

I recently added Opencore Gui and Bootchime to my OC 0.5.8 on 

Board: Gigabyte Z170X-UD5 TH
i7 6700 K
Sapphire RX580 8GB
Original Apple Extreme on PCi adaptor

Samsung 970 %00GB

After updating settings I reboot and clear NVRAM via the selection in the GUI

and I get bootchime and all the correct icons in the picker using the esc key. It boots to OS and when I set the startup drive in Preferences and then reboot I lose the access to the picker and lose the bootchime. I just boots straight to my startup drive. If I reset picker to Builtin I can reset NVRAM and the same problem is repeatable. I cant figure out why it loses the settings.

 

Ive included my config.plist

config.plist

[kocoman]

Does anyone know what it means by 

• Added boot.efi debug protocol support for 10.15.4+
• Added boot.efi performance logging for 10.15.4+

thanks

[Sander Spilleman]

2 hours ago, vit9696 said:

1. ProtocolOverrides — you do not need to override AppleRtcRam as you do not have Apple hardware. Leave it as false.

2. wake-failure can and should be removed.

3. AppleRTC kernel patches can be removed (they are replaced by a quirk).

4. You forgot to mention rtc-blacklist, but I should remind you that it needs to be set to 58 59 for Z390-like.

5. I don't know whether you need an ACPI patch.

 

As for ACPI patches and

 

Thank you so much @vit9696. I'll make the changes!

[shhbz]

Hi 

I wonder if someone can share some idea on what's wrong on my Catalina built.

Issue 1. Catalina on Opencore 0.5.8 does not restart or shutdown, it blinks the screen once and stays on the homescreen Catalina Dekstop.

Issue 2. PROXITY_WAKE=1 so how to change to 0 as i did it with terminal command 

sudo pmset -a proximitywake 0

but on reboot it reset itself back to 1, so when I try to wake the pc from sleep using wireless mouse mx master 2s it does not wake up.the CPU turns on but not the screen.

 

Config: MSI Z170krait Gaming3x ,6700k Skylake, 32gb Gskill, Sapphire Rx 580

 

Attaching below my Opencore EFI and Energy Saver settings 

EFI.zip

Edited May 7, 2020 by shhbz

[Synapsis]

I wasted the whole day and still can't make OC work...

 

I'm running a z370 / i-8700K / RX Vega 56, I wasn't using my box for a couple of months and now 

not matter which freaking config I use I can't get past the login screen which hangs and macOS reboots.

 

- I added HPET/TMR/IRQ releated patches through SSDTime util

- I use newest .kext builds

 

I need the iGPU, so don't tell me why/don't. Period.

I've got a ASUS Prime z370-P cheap-ass motherboard, 32GB ram, i7-i8700K and a ASROCK RX Vega gaming something flashed with 64 bios.

I got my own platform/nvram settings which aren't in the config plist since I kind of like those serial numbers. 

What happens:

 

Freeze on line: 

Quote

IOConsoleUsers: gIOScreenLockState 3, hs 0,bs 0, now 0, sm 0x0

 

Then it flips to the VEGA DP output and then it freezes again, gets all choppy and then freezes forever or crashes,

I can look at the Catalina picture and {censored} off to it... 

 

It certainly is an issue with IGPu which I use for driving a 3rd screen, and which wasn't a problem until, well, now

After reading docs im more confused then I was before - since I don't know which keys are actually proper to use and which not.

For example - docs says I don't need 'shikigva=128' but without this flag it won't boot to nothing, no matter which config...

Also, there is AAPL,ig-platform-id and device-id, I get it, but - is it AAPL,device-id or device-id in the key's name?

Also, how can I calculate those hex values Xcode wraps the values around in .plist without Xcode? 

I'm running on Linux, so I vim around the config.

Anyways, enough whining.

 

Please help a fellow hackintosher - Ill spend you a beer!

I'm rising my hair out and may need a aggression management therapy after today.

 

My (newly edited) config.plist is attached.

 

I dont believe I can't make this thing work since it worked perfectly without a hitch until, well, Catalina... 

f**k me to tears.

 

SIDENOTE: I used 0.5.9 upstream sources and builded it on linux, it compiles without a hitch, but since a few commits- packages build but the binary wont load. hmm 

  

config.plist

[canyondust]

This might be a silly question, but I believe the people here know the answer: my firmware has errors in SSDT-5. If I fix those errors and place SSDT-5 in EFI/OC/ACPI/ and add it to my config, will it supercede the firmware SSDT-5?

 

Thanks and cheers!

 

Edit: documentation provides the answer. always RTFM.

Edited May 8, 2020 by canyondust
I read TFM.

[Alex009988]

Hi. Is it possible to use opencore with HFS+ on mbr?  The problem is it doesn't see the macos partition or I can't see necessary hfsplusefi driver.

[shhbz]

On 5/4/2020 at 6:49 PM, Maf3r0 said:

 

 

changes that help-me a lot:

•  
• Added ResetSystem tool to allow shutdown/reset actions in the menu

where to find this option, as unable to find it

On 5/4/2020 at 6:49 PM, Maf3r0 said:

 

 

[Matgen84]

18 minutes ago, shhbz said:

where to find this option, as unable to find it

 

 

You have to add ResetSystem.efi to Tools section in your config.plist. Copy/past the driver in EFI/OC/Tools. I suppose. 

[genzai]

feature request:

Can SMBIOS data be separated from config.plist (in separate smbios.plist for instance)? separating this out would make EFI much more (fully?) portable. people could post working EFI or config.plist and SMBIOS data will always be secure, persistent, and easy to add and maintain separately.

[Cloudane]

On 3/29/2020 at 6:02 PM, Hoshi Yamazaki said:

Ok, so I got almost working everything on my x220 (even brightness control and battery status!) but for some reason VoodooPS2/VoodooInput is not loading, and I tried acidanthera version and also one from T420 thread and I'm not sure what is wrong.

Any advice? It /was/ working previously on Clover so I assume this is OpenCore quirk. Maybe kexts order?

 

 

MFW I'm still struggling to get audio working on my X220.  Tried alcid 3, 12, 13, 14 and 28 as the codec is allegedly compatible with CX20590

 

Edit: SUCCESS (albeit without CPU power management)

tldr I had to fix HPET.  This then caused a kernel panic saying CPU 3 doesn't have a HPET, breaking the work I'd previously done on getting CPU power management working - but after enabling the DummyPowerManagement quirk I have a booting system with audio. 

 

For whatever reason when creating HPET patch with SSDTTool I didn't put the patches into the config.plist for opencore.  This was kind of annoying as at this point I didn't have Windows on the machine any more and had to produce the patch with a Linux live USB, because SSDTTool doesn't extract the DSDT on MacOS.  What followed after that was hours of wrangling with bugs in SSDTTool for Linux, finding a workaround tool called FixHPET and wrangling with bugs in that too... but eventually I got my SSDT patch and config.plist patches.

Edited May 8, 2020 by Cloudane

[gorans]

On 5/4/2020 at 5:04 PM, kiu777 said:

The problem is that my external SSD which I use as an external clone of my Mac OS boot drive is recognized as internal disk

despite being connected via USB3. If I then throw this clone out the hidden system volume is still mounted and I get the

typical error messages from Mac OS that I disconnected a drive without proper dismounting.

Hi Kiu,

did You manage to solve this problem? I also have USB flash drives listed as "USB Internal Physical Disk" but they have external icons and I can unmount them.